Article

Judgement of Act-Outcome Contingency: The Role of Selective Attribution

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

In the first experiment subjects were presented with a number of sets of trials on each of which they could perform a particular action and observe the occurrence of an outcome in the context of a video game. The contingency between the action and outcome was varied across the different sets of trials. When required to judge the effectiveness of the action in controlling the outcome during a set of trials, subjects assigned positive ratings for a positive contingency and negative ratings for a negative contingency. Furthermore, the magnitude of the ratings was related systematically to the strength of the actual contingency. With a fixed probability of an outcome given the action, judgements of positive contingencies decreased as the likelihood that the outcome would occur without the action was raised. Correspondingly, the absolute value of ratings of negative contingencies was increased both by an increment in the probability of the outcome in the absence of the action and by a decrement in the probability of the outcome following the action. A systematic bias was observed, however, in that positive judgements were given under a non-contingent relationship when the outcome frequency was relatively high. However, this bias could be reduced by giving extended exposure to the non-contingent schedule (Experiment 2). This pattern of contingency judgements can be explained if it is assumed that a process of selective attribution operates, whereby people are less likely to attribute an outcome to some potential target cause if another effective cause is present. Experiments 2 and 3 demonstrated the operation of this process by showing that initially establishing another agent as an effective cause of the outcome subsequently reduced or blocked the extent to which the subjects attributed the outcome to the action. Finally, we argue that the pattern and bias in contingency judgements based upon interactions with a causal process can be explained in terms of contemporary conditioning models of associative learning.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Moreover, a wealth of data has shown that instrumental actions in humans can also be sensitive to the same causal manipulation and that, more importantly, human causal beliefs of the effectiveness of an action in producing a reward correlate with instrumental performance. Shanks and Dickinson (1991), for example, probed participants' causal beliefs by presenting them with a contingency in which the probability of obtaining a reward after each key press was fixed and manipulated the probability of obtaining and reward in the absence of a key press and found that the causal ratings followed the difference between these two probabilities (Dickinson, Shanks, & Evenden, 1984;Hammond, 1980;Vaghi et al., 2019;Wasserman, Chatlosh, & Neunaber, 1983); instrumental performance as measured by rate of key pressing also followed the same difference. ...
... Procedure. Following previous experimental procedures in humans (Dickinson et al., 1984;Reed, 2001;Shanks & Dickinson, 1991;Wasserman et al., 1983), in each trial participants were presented with a distinguishable background color that signalled which reward schedule, interval or ratio, was in effect. Each condition was presented for 3 minutes. ...
... Data, manuscript and analyses for this paper can be found in https://osf.io/b7re6/. These results suggest a dissociation between instrumental performance and causal beliefs under free-operant training and are consistent with the widely-held view that causality is attributed mainly on the basis of the probability of reward per action (Chatlosh, Neunaber, & Wasserman, 1985;Cheng, 1997;Cheng & Buehner, 2012;Dickinson et al., 1984;Shanks & Dickinson, 1991;Wasserman et al., 1983). What is not clear, however, is the role that the action-reward rate correlation process would play in increasing action rates on the ratio over the interval schedule in spite of similar experienced reward probabilities. ...
Article
Full-text available
Human experiments have demonstrated that instrumental performance of an action and the causal beliefs of the effectiveness of an action in producing a reward are correlated and controlled by the probability of an action leading to a reward. The animal literature, however, shows that instrumental performance under free-operant training differs even when these reward probabilities are matched while subjects undergo training under ratio or interval schedules of reward. In two experiments, we investigated whether causal beliefs would correlate with instrumental performance under interval and ratio schedules for matched reward probabilities. In both experiments we found that performance was higher under ratio than under interval training. However, causal beliefs were similar between these two conditions despite these differences in instrumental performance. When reward probabilities were increased by experimental manipulations in Experiment 2, the causal beliefs increased but performance decreased with respect to Experiment 1. This is evidence that instrumental performance and causal action-reward attribution may not follow from the same psychological process under free-operant training.
... The discussion of Agent's Paradox is a fundamental problem for the understanding of people's behavior. Previous studies have reported experimental analysis of causation learning in terms of action-outcome (A-O) contingency (Allan, 1980;Dickinson, Shanks, & Evenden, 1984;Wasserman, Elek, Chatlosh, & Baker, 1993) [4][5][6]. A-O Contingency refers to the relative probability of the outcome's occurrence in the presence of a goal-directed action minus its probability in the absence of the action. ...
... The discussion of Agent's Paradox is a fundamental problem for the understanding of people's behavior. Previous studies have reported experimental analysis of causation learning in terms of action-outcome (A-O) contingency (Allan, 1980;Dickinson, Shanks, & Evenden, 1984;Wasserman, Elek, Chatlosh, & Baker, 1993) [4][5][6]. A-O Contingency refers to the relative probability of the outcome's occurrence in the presence of a goal-directed action minus its probability in the absence of the action. Learning about the sense of agency in different action-outcome contingencies is crucial for people to take appropriate action in a dynamic environment. ...
... Chapman and Chapman (1967) found a tendency to overestimate the casual relativity between symptoms and diagnoses in zero or low contingencies due to the high outcome density [8]. The same bias is also shown in A-O contingency judgment: though people rate a high control of actions over outcome in high A-O contingency conditions and rate a low control in low A-O contingency conditions, they also have strong tendency to evaluate a higher degree of control in high-density zero contingency condition ( [5,[9][10][11][12][13]. Together, these studies highlight a bias of control judgment based on specific A-O contingency information. ...
Article
By action-outcome causality learning people are able to gain a sense of agency which motivates people to take goal-directed actions. However, people' sensation of agency can be influenced by the multi-agent context where the causation between agents and outcome are complex. The present study examine how people judge the sense of agency of themselves and others when presented with different contingencies of the outcome controlled by the individual and another agent. The results shows that People can learn action-outcome causation from action-effect causal perception in multi-agent context, gain sense of agency when they do have control on the outcome and learn the agency of other agent when other agent has the causal power on the outcome. The present study is an step forward to understand how individuals work and perceive sense of agency with a external agent, which may give a hint to further inquiry of computational modeling and analysis of brain activities in the development of agency.
... This view of causation and prevention as diametrically opposed is formalised in traditional associative learning models such as the Rescorla-Wagner model (Rescorla & Wagner, 1972). These models, which were originally developed to explain Pavlovian conditioning in animals, have been cited as theoretical accounts of how humans learn causal and preventive relationships between cues and outcomes (Dickinson et al., 1984). The Rescorla-Wagner model conceptualises predictive relationships between cues and outcomes in terms of a single dimension of associative strength. ...
... The causal learning literature shows that participants interpret associations differently according to the causal model implied by the cover story (Waldmann et al., 2006;Waldmann & Holyoak, 1992), but there have been few attempts to investigate whether different causal models (structures) arise from associative learning procedures with more neutral cover stories. Such a connection would further strengthen the view that associative learning theories can account for how humans learn about causal relationships (Dickinson et al., 1984) and help clarify the role of higher-order cognitive processes in human associative learning Lee, Lovibond, Hayes, & Navarro, 2019;McLaren et al., 2018;Mitchell, De Houwer, & Lovibond, 2009). ...
... We assessed causal structure in two ways. The first was via open-ended self-report and the second was with a three-alternative forced-choice (3AFC) question with options describing a preventive, modulatory, and configural causal structure for B. We also included a standard causal ratings test (e.g., Dickinson et al., 1984) to assess what participants learned about the causal status of each individual cue on a bipolar cause-prevent scale. Note, however, that our main hypothesis involved testing whether causal structure predicted differences in the transfer of inhibitory learning in the outcome prediction test. ...
Article
Full-text available
Traditional associative learning theories predict that training with feature negative (A+/AB-) contingencies leads to the feature B acquiring negative associative strength and becoming a conditioned inhibitor (i.e., prevention learning). However, feature negative training can sometimes result in negative occasion setting, where B modulates the effect of A. Other studies suggest that participants learn about configurations of cues rather than their individual elements. In this study, we administered simultaneous feature negative training to participants in an allergist causal learning task and tested whether evidence for these three types of learning (prevention, modulation, configural) could be captured via self-report in the absence of any procedural manipulation. Across two experiments, we show that only a small subset of participants endorse the prevention option, suggesting that traditional associative models that predict conditioned inhibition do not completely capture how humans learn about negative contingencies. We also show that the degree of transfer in a summation test corresponds to the implied causal structure underlying conditioned inhibition, occasion-setting, and configural learning, and that participants are only partially sensitive to explicit hints about causal structure. We conclude that feature negative training is an ambiguous causal scenario that reveals individual differences in the representation of inhibitory associations, potentially explaining the modest group-level inhibitory effects often found in humans.
... It is generally assumed that the Rescorla and Wagner (1972) model adequately accommodates the full results of simple cue competition experiments in humans (e.g. Dickinson et al., 1984), while the Bush and Mosteller (1951) model cannot. We present simulations that demonstrate this assumption is wrong in at least some circumstances. ...
... The model can account for blocking because A takes up all the available associative strength on the A+ trials, meaning that there is no available associative strength for X to acquire on the AX+ trials (assuming learning is at asymptote). Dickinson et al. (1984) provided one of the first experimental demonstrations of blocking in humans. The Rescorla and Wagner (1972) model is widely assumed to adequately explain the full data of such experiments, whilst also providing a better account than Bush and Mosteller's (1951) model. ...
... There are several simple human forward cue-competition experiments reported in the literature (e.g. Dickinson et al., 1984;Miller, 1996;Mitchell & Lovibond, 2002), but none of these datasets have been made openly accessible. In order to provide an openly accessible dataset, we ran a standard forward cue competition experiment. ...
Article
Full-text available
It is generally assumed that the Rescorla and Wagner (1972) model adequately accommodates the full results of simple cue competition experiments in humans (e.g. Dickinson et al., 1984), while the Bush and Mosteller (1951) model cannot. We present simulations that demonstrate this assumption is wrong in at least some circumstances. The Rescorla-Wagner model, as usually applied, fits the full results of a simple forward cue-competition experiment no better than the Bush-Mosteller model. Additionally, we present a novel finding, where letting the associative strength of all cues start at an intermediate value (rather than zero), allows this modified model to provide a better account of the experimental data than the (equivalently modified) Bush-Mosteller model. This modification also allows the Rescorla-Wagner model to account for a redundancy effect experiment (Uengoer et al., 2013); something that the unmodified model is not able to do. Furthermore, the modified Rescorla-Wagner model can accommodate the effect of varying the proportion of trials on which the outcome occurs (i.e. the base rate) on the redundancy effect (Jones et al., 2019). Interestingly, the initial associative strength of cues varies in line with the outcome base rate. We propose that this modification provides a simple way of mathematically representing uncertainty about the causal status of novel cues within the confines of the Rescorla-Wagner model. The theoretical implications of this modification are discussed. We also briefly introduce free and open resources to support formal modelling in associative learning. Keywords: associative learning, prediction error, uncertainty, modelling, blocking, redundancy effect, open science.
... With respect to action-outcome contingency, several experiments conducted mainly in the 1980s and early 1990s (e.g. Allan & Jenkins, 1980Alloy & Abramson, 1979;Chatlosh, Neunaber & Wasserman, 1985;Dickinson, Shanks & Evenden, 1984;Neunaber & Wasserman, 1986;Shanks, 1987Shanks, , 1989Shanks & Dickinson, 1991;Shanks, Pearson & Dickinson, 1989;Wasserman, Chatlosh & Neunaber, 1983) demonstrated that action-outcome contingency was an important determinant of how much control participants considered they had over action outcomes. For example, Shanks and Dickinson (1991) tested several groups of participants, with each group able to cause an outcome to occur with a set probability (.85) by pressing a button. ...
... Thus, these findings are consistent with previous research showing that participants are sensitive to changes in action outcome contingency (e.g. Allan & Jenkins, 1980Alloy & Abramson, 1979;Chatlosh et al., 1985;Dickinson et al., 1984;Neunaber & Wasserman, 1986;Shanks, 1987Shanks, , 1989Shanks & Dickinson, 1991;Shanks et al., 1989;Sidarus et al., 2013;Wasserman et al., 1983). ...
Article
Full-text available
Tasks measuring the sense of agency often manipulate the predictability of action outcomes by introducing spatial deviation. However, the extent to which spatial predictability of an outcome influences the sense of agency when spatial deviation is controlled for remains untested. We used a novel task to investigate the effect of several factors (action-outcome contingency, spatial deviation, and spatial predictability when controlling for spatial deviation of action outcomes) on the sense of agency. We also investigated trait predictors of metacognition of agency-the degree to which participants' confidence in their agency judgements corresponds to the accuracy of those judgements. Initial and replication samples completed Contingency, Deviation, and Predictability versions of the task. Across samples, participants' sense of agency was impacted by action-outcome contingency and spatial deviation of action outcomes. Manipulation of the spatial predictability of action outcomes did not reliably impact the sense of agency. Metacognition of agency was related to alexithymic traits-higher alexithymia scores were associated with reduced metacognition of agency.
... However, there is also widespread evidence that humans can be highly sensitive to the objective contingencies between actions and their consequences. For example, explicit judgements of causality (Dickinson et al., 1984) and agency (Sato, 2009), along with implicit markers of action-outcome learning (Elsner & Hommel, 2004), increase as contingencies between actions and events become stronger. Similar patterns are seen in 'intentional binding' experiments, where shifts in subjective time perception are used as an implicit marker of the sense of agency (Haggard, Clark, & Kalogeras, 2002;Moore & Haggard, 2008;Moore, Lagnado, Deal, & Haggard, 2009). ...
... These results demonstrate that illusions of control may not imply grandiose beliefs about our capabilities that disregard objective evidence (Alloy & Abramson, 1979;Presson & Benassi, 2003). They in fact allow us to conceptualise illusions of control in line with other findings where we are highly sensitive to variations in the contingency between our actions and their outcomes (Dickinson et al., 1984;Elsner & Hommel, 2004;Moore & Haggard, 2008;Moore et al., 2009). They are also consistent with recent findings from perceptual decision making where biases to report seeing signals in noise can reflect spurious signals rather than complete hallucinations (Wyart et al., 2012). ...
Preprint
Full-text available
We frequently experience feelings of agency over events we do not objectively influence – so-called ‘illusions of control’. These illusions have prompted widespread claims that we can be insensitive to objective relationships between actions and outcomes, and instead rely on grandiose beliefs about our abilities. However, these illusory biases could instead arise if we are highly sensitive to action-outcome correlations, but attribute agency when such correlations emerge simply by chance. We motion-tracked participants while they made agency judgements about a cursor that could be yoked to their actions or follow an independent trajectory. A combination of signal detection analysis, reverse correlation methods and computational modelling indeed demonstrated that ‘illusions’ of control could emerge solely from sensitivity to spurious action-outcome correlations. Counterintuitively, this suggests that illusions of control could arise because agents have excellent insight into the relationships between actions and outcomes in a world where causal relationships are not perfectly deterministic.
... However, there is also widespread evidence that humans can be highly sensitive to the objective contingencies between actions and their consequences. For example, explicit judgements of causality (Dickinson et al., 1984) and agency (Sato, 2009), along with implicit markers of action-outcome learning (Elsner & Hommel, 2004), increase as contingencies between actions and events become stronger. Similar patterns are seen in 'intentional binding' experiments, where shifts in subjective time perception are used as an implicit marker of the sense of agency (Haggard et al., 2002;Moore et al., 2009; explanation for illusions of control -that these experiences arise because agents are especially sensitive to relationships between actions and outcomes and pick up on those that occur purely by chance. ...
... These results demonstrate that illusions of control may not imply grandiose beliefs about our capabilities that disregard objective evidence (Alloy & Abramson, 1979;Presson & Benassi, 2003). They in fact allow us to conceptualise illusions of control in line with other findings where we are highly sensitive to variations in the contingency between our actions and their outcomes (Dickinson et al., 1984;Elsner & Hommel, 2004;Moore et al., 2009;Moore & Haggard, 2008). They are also consistent with recent findings from perceptual decision making where biases to report seeing signals in noise can reflect spurious signals rather than complete hallucinations (Wyart et al., 2012). ...
Article
Full-text available
We frequently experience feelings of agency over events we do not objectively influence - so-called 'illusions of control'. These illusions have prompted widespread claims that we can be insensitive to objective relationships between actions and outcomes, and instead rely on grandiose beliefs about our abilities. However, these illusory biases could instead arise if we are highly sensitive to action-outcome correlations, but attribute agency when such correlations emerge simply by chance. We motion-tracked participants while they made agency judgements about a cursor that could be yoked to their actions or follow an independent trajectory. A combination of signal detection analysis, reverse correlation methods and computational modelling indeed demonstrated that 'illusions' of control could emerge solely from sensitivity to spurious action-outcome correlations. Counterintuitively, this suggests that illusions of control could arise because agents have excellent insight into the relationships between actions and outcomes in a world where causal relationships are not perfectly deterministic.
... In a recent human study, Pérez and colleagues (2016) (Blanco, Matute, & Vadillo, 2013;Dickinson, Shanks, & Evenden, 1984;Hammond & Paynter, 1983;Wasserman, Chatlosh, & Neunaber, 1983) or variations of this metric (Cheng, 1997;Novick & Cheng, 2004). The most plausible explanation for Reed's (2001) results is that participants attribute causal control in line with the correlational properties of the schedules, independently of reward rate or reward probability. ...
Preprint
Full-text available
The higher response rates observed on ratio than on matched interval reward schedules has been attributed to the differential reinforcement of longer inter-response times (IRTs) on the interval contingency. Some data, however, seem to contradict this hypothesis, showing that the difference is still observed when the role of IRT reinforcement is neutralized by using a regulated-probability interval schedule (RPI). Given the mixed evidence for these predictions, we re-examined this hypothesis by training three groups of rats to lever press under ratio, interval and RPI schedules across two phases while matching reward rates within triads. At the end of the first phase, the master ratio and RPI groups responded at similar rates. In the second phase, an interval group yoked to the same master ratio group of the first phase responded at a lower rate than the RPI group. Post-hoc analysis showed comparable reward rates for master and yoked schedules. The experienced response-outcome rate correlations were likewise similar, and approached zero as training progressed. We discuss these results in terms of dual-system theories of instrumental conditioning.
... It was less clear whether cue competition can be found in incidental learning tasks, that is, tasks in which predictive cues were not task relevant (targets). In most past reports, learning the contingency was the explicit goal and participants had ample time to reflect on the events that they saw (Chapman & Robbins, 1990;Dickinson, Shanks, & Evenden, 1984;Gluck & Bower, 1988;Le Pelley & McLaren, 2001). Thus, in our report we utilised the colour-word contingency learning paradigm for studying overshadowing and blocking. ...
... Blocking, overexpectation, and other cue competition effects have been demonstrated in a variety of species ranging from honeybees (e.g., Couvillon & Bitterman, 1989) to humans (e.g., Dickinson, Shanks, & Evenden, 1984), and across different experimental procedures, including aversive (e.g., Wagner, Logan, Haberlandt, & Price, 1968), appetitive (e.g., Lattal & Nakajima, 1998), and neutral (e.g., Melchers, Üngör, & Lachnit, 2005) learning tasks. However, there have also been reliable failures to find evidence for competition among cues (e.g., Maes et al., 2016). ...
Article
Full-text available
Cue competition refers to phenomena indicating that learning about the relationship between a cue and an outcome is influenced by learning about the predictive significance of other cues that are concurrently present. In two autoshaping experiments with pigeons, we investigated the strength of competition among cues for predictive value. In each experiment, animals received an overexpectation training (A+, D+ followed by AD+). In addition, the training schedule of each experiment comprised two control conditions-one condition to evaluate the presence of overexpectation (B+ followed by BY+) and a second one to assess the strength of competition among cues (C+ followed by CZ-). Training trials were followed by a test with individual stimuli (A, B, C). Experiment 1 revealed no evidence for cue competition as responding during the test mirrored the individual cue-outcome contingencies. The test results from Experiment 2, which included an outcome additivity training, showed cue competition in form of an overexpectation effect as responding was weaker for Stimulus A than Stimulus B. However, the test results from Experiment 2 also revealed that responding to Stimulus A was stronger than to Stimulus C, which indicates that competition among cues was not as strong as predicted by some influential theories of associative learning.
... In a blocking scenario, for example, participants experience A followed by an outcome over a number of trials and are subsequently presented with the compound AB followed by the same outcome over a comparable number of trials. The fact that the causal rating for B is usually lower than what it would have been had it not been trained in compound with a previously trained A is regarded as evidence of A blocking the acquisition of causal strength to B (Dickinson, Shanks, & Evenden, 1984). In an experiment reported by Livesey and Boakes (2004, Experiment 3), a compound of two stimuli combined into a single object significantly reduced blocking to B compared to a condition in which the two cues were presented separated on the screen. ...
Article
Full-text available
Several contemporary models anticipate that the summation effect is modulated by the similarity between the cues forming a compound. Here, we explore this hypothesis in a series of causal learning experiments. Participants were presented with two visual cues that separately predicted a common outcome and later asked for the outcome predicted by the compound of the two cues. Similarity was varied between groups through changes in shape, spatial position, color, configuration, and rotation. In variance with the predictions of these models, we observed similar and strong levels of summation in both groups across all manipulations of similarity. The effect, however, was significantly reduced by manipulations intended to impact assumptions about the causal independence of the cues forming the compound, but this reduction was independent of stimulus similarity. These results are problematic for similarity-based models and can be more readily explained by rational approaches to causal learning.
... Associatieve theorieën over menselijk leren kennen een lange geschiedenis (Dickinson, Shanks, & Evenden, 1984) en bouwen voort op de theorieën van associatief leren bij dieren (Pavlov, 1927). De kern van associatieve theorieën komt hierop neer: wanneer een stimulus herhaaldelijk met een bepaalde uitkomst wordt aangeboden, ontwikkelen er zich in ons geheugen mentale representaties van beide waartussen een associatie (een verbinding) ontstaat. ...
... Although such tasks differ from conditioning paradigms in various ways (e.g., no biologically salient outcome, explicit predictions measured rather than conditioned behaviour), predictive learning tasks are commonly assumed to measure associative learning processes (De Houwer & Beckers, 2002;Shanks, 1985;Wasserman & Miller, 1997). Many of the same classic phenomena displayed in animal studies such as blocking (Dickinson, Shanks, & Evenden, 1984) and conditioned inhibition (Lee & Livesey, 2012) replicate in predictive learning paradigms. Thus, behaviour in these tasks is readily explained by appealing to conditioning theories and assuming that associative strength maps onto explicit outcome predictions, without the need to appeal to higher order reasoning. ...
Article
In property induction tasks, encountering a diverse range of instances (e.g., hippos and hamsters) with a given property usually increases our willingness to generalise that property to a novel instance, relative to non-diverse evidence (e.g., hippos and rhinos). Although generalisation in property induction and predictive learning tasks share conceptual similarities, it is unknown whether this diversity principle applies to generalisation of a predictive association. We tested this hypothesis in two predictive learning experiments using differential training where one category of stimuli (e.g., fruits) predicted an outcome and another category (e.g., vegetables) predicted no outcome. We compared generalisation between a Non-Diverse group who were presented with non-diverse evidence in both positive (predicted the outcome) and negative (predicted no outcome) categories, and two groups who received the same training as the Non-Diverse group but with a more diverse range of exemplars in the positive (Diverse+ group) or negative (Diverse- group) category. Diversity effects were found for both positive and negative categories, in that learning about a diverse range of exemplars increased generalisation of a predictive association to novel exemplars from that same category. The results suggest that diversity, a key principle describing how we reason inductively, also applies to generalisation in associative learning tasks.
... Participants also rated their expectancy of shock on each trial during the CS presentation period. Associative learning theorists have proposed that self-reported expectancy ratings reflect the learning of an association between CS and US over trials (Dickinson, Shanks & Evenden, 1984;Shanks, 2007). As such, while expectancy ratings are likely to involve symbolic cognitive processes, they are also taken to reflect associatively-guided learning processes. ...
Article
Two experiments explored the role of verbalisable rules in generalisation of human differential fear conditioning with electric shock as the aversive stimulus. Two circles of different sizes served as conditioned stimuli (CS+ and CS-), before testing with a range of circle sizes. In Experiment 1, shock expectancy ratings followed a peak-shifted unimodal gradient, with maximum ratings at a test value further along the dimension from CS+ in the opposite direction to CS-. However, differentiable gradients were observed when participants were divided on the basis of the rules they reported using during the task (linear and similarity). Experiment 2 was designed to counter the contradictory feedback arising from extinction testing by removing the shock electrodes during the test phase. A more linear overall gradient was observed, and sub-groups defined by self-reported rules showed distinct gradients that were congruent with their rules. These results indicate that rule-based processes are influential in generalisation of conditioned fear along simple stimulus dimensions, and may help explain generalisation phenomena that have traditionally been attributed to automatic, similarity-based processes.
... • The blocking effect in human contingency learning is consistent with this as it suggests that people fail to encode an association between a cue and outcome if that outcome is already expected on the basis of other predictive cues (Dickinson, Shanks & Evenden, 1984). ...
Poster
Full-text available
To what extent are the predictions that are important for human learning under flexible control?
... Although such tasks differ from conditioning paradigms in various ways (e.g., no biologically salient outcome, explicit predictions measured rather than conditioned behavior), predictive learning tasks are commonly assumed to measure associative learning processes (De Houwer & Beckers, 2010;Shanks, 1985;Wasserman & Miller, 1997). Many of the same classic phenomena displayed in animal studies such as blocking (Dickinson, Shanks, & Evenden, 1984) and conditioned inhibition (Lee & Livesey, 2012) replicate in predictive learning paradigms. Thus, behavior in these tasks is readily explained by appealing to conditioning theories and assuming that associative strength maps onto explicit outcome predictions, without the need to appeal to higher-order reasoning. ...
Article
Full-text available
In property induction tasks, encountering a diverse range of instances (e.g., hippos and hamsters) with a given property usually increases our willingness to generalize that property to a novel instance, relative to non-diverse evidence (e.g., hippos and rhinos). Although generalization in property induction and predictive learning tasks share conceptual similarities, it is unknown whether this diversity principle applies to generalization of a predictive association. We tested this hypothesis in two predictive learning experiments using differential training where one category of stimuli (e.g., fruits) predicted an outcome and another category (e.g., vegetables) predicted no outcome. We compared generalization between a Non-Diverse group who were presented with non-diverse evidence in both positive (predicted the outcome) and negative (predicted no outcome) categories, and two groups who received the same training as the Non-Diverse group but with a more diverse range of exemplars in the positive (Diverse + group) or negative (Diverse - group) category. Diversity effects were found for both positive and negative categories, in that learning about a diverse range of exemplars increased generalization of a predictive association to novel exemplars from that same category. The results suggest that diversity, a key principle describing how we reason inductively, also applies to generalization in associative learning tasks.
... One issue that still remains unclear is whether cue competition can be found in incidental learning tasks, that is, tasks in which predictive cues were not task relevant (targets). In most past reports, learning the contingency was the explicit goal and participants had ample time to reflect on the events that they saw (Chapman & Robbins, 1990;Dickinson, Shanks, & Evenden, 1984;Gluck & Bower, 1988;D. R. Shanks, 1985). ...
Article
Full-text available
Overshadowing and blocking are two important findings that are frequently used to constrain models of associative learning. Overshadowing is the finding that learning about a cue (referred to as X) is reduced when that cue is always accompanied by a second cue (referred to as A) during the learning phase (AX). Blocking is the finding that after learning a stimulus-outcome relation for one stimulus (A), learning about a second stimulus (X) is reduced when the second stimulus is always accompanied by the first stimulus (AX). It remains unclear whether overshadowing and blocking result from explicit decision processes (e.g., "I know that A predicts the outcome, so I am not sure whether X does, too"), or whether cue competition is built directly into low-level association formation processes. In that vein, the present work examined whether overshadowing and/or blocking are present in an incidental learning procedure, where the predictive stimuli (words or shapes) are irrelevant to the cover task and merely correlated with the task-relevant stimulus dimension (colour). In two large online studies, we observed no evidence for overshadowing or blocking in this setup: (a) no evidence for an overshadowing cost was observed with compound (word-shape) cues relative to single cue learning conditions, and (b) contingency learning effects for blocked stimuli did not differ from those for blocking stimuli. However, when participants were given the explicit instructions to learn contingencies, evidence for blocking and overshadowing was observed. Together, these results suggest that contingencies of blocked/overshadowed stimuli are learned incidentally, but are suppressed by explicit decision processes due to knowledge of the contingencies for the blocking/overshadowing stimuli.
... Soon after it was published, the Rescorla-Wagner model began to also be effectively applied to human learning in areas as far-ranging as paired-associate word learning, category learning, correlational relationship judgments, transitive inference reasoning, social psychology, visual perception and physiological regulation (reviewed in Siegel & Allan, 1996). Feedback from prediction error also plays a vital role in the development of motor control (Cheng & Sabes, 2007;Shadmehr, Smith, & Krakauer, 2010) and human contingency learning (Dickinson, Shanks, & Evenden, 1984;Houwer & Beckers, 2002), as well as time perception (Ramscar, Matlock, & Dye, 2010) and pitch perception in music (Ramscar, Suh, & Dye, 2011). There is also neuroscientific evidence for the role of error in learning (see Schultz, 1998, for a review). ...
Article
Full-text available
Despite burgeoning evidence that listeners are highly sensitive to statistical distributions of speech cues, the mechanism underlying learning may not be purely statistical tracking. Decades of research in animal learning suggest that learning results from prediction and prediction error. Two artificial language learning experiments test two predictions that distinguish error-driven from purely statistical models; namely, cue competition – specifically, Kamin’s (1968) ‘blocking’ effect (Experiment 1) – and the predictive structure of learning events (Experiment 2). In Experiment 1, prior knowledge of an informative cue blocked learning of a second cue. This finding may help explain second language learners’ difficulty in acquiring native-level perception of non-native speech cues. In Experiment 2, learning was better with a discriminative (cue–outcome) order compared to a non-discriminative (outcome–cue) order. Experiment 2 suggests that learning speech cues, including reversing effects of blocking, depends on (un)learning from prediction error and depends on the temporal order of auditory cues versus semantic outcomes. Together, these results show that (a) existing knowledge of acoustic cues can block later learning of new cues, and (b) speech sound acquisition depends on the predictive structure of learning events. When feedback from prediction error is available, this drives learners to ignore salient non-discriminative cues and effectively learn to use target cue dimensions. These findings may have considerable implications for the field of speech acquisition.
... Then, during the subsequent training phase, the context would have been a relatively strong competitor of the target cue, leading to lower causal judgments in the pretrained group than in the Control group. That is, the context, given its previous association with the outcome, could block further learning about the target potential cause [40]. Thus, blocking is a commonly observed learning phenomenon, predicted by the associative theory, that can easily explain what happened in Experiment 1. ...
Article
Full-text available
Previous research revealed that people’s judgments of causality between a target cause and an outcome in null contingency settings can be biased by various factors, leading to causal illusions (i.e., incorrectly reporting a causal relationship where there is none). In two experiments, we examined whether this causal illusion is sensitive to prior expectations about base-rates. Thus, we pretrained participants to expect either a high outcome base-rate (Experiment 1) or a low outcome base-rate (Experiment 2). This pretraining was followed by a standard contingency task in which the target cause and the outcome were not contingent with each other (i.e., there was no causal relation between them). Subsequent causal judgments were affected by the pretraining: When the outcome base-rate was expected to be high, the causal illusion was reduced, and the opposite was observed when the outcome base-rate was expected to be low. The results are discussed in the light of several explanatory accounts (associative and computational). A rational account of contingency learning based on the evidential value of information can predict our findings.
... A wealth of data have shown 48 that instrumental actions in humans can also be sensitive to the same causal manipulation 49 and that, more importantly, human causal beliefs of the effectiveness of an action in 50 producing an outcome correlate with instrumental performance. Shanks and Dickinson 51 (1991), for example, probed participants' causal beliefs by presenting them with a 52 contingency in which the probability of obtaining an outcome after each key press was fixed 53 and manipulated the probability of obtaining and outcome in the absence of a key press and 54 found that the causal ratings followed the difference between these two probabilities: 55 ∆P = P (O|A) − P (O|¬A) (Dickinson, Shanks, & Evenden, 1984;Hammond, 1980;Vaghi et 56 al., 2018;Wasserman, Chatlosh, & Neunaber, 1983); instrumental performance as measured 57 by rate of key presses also followed the same metric. Although the question of whether these 58 causal beliefs are reflecting A-O causality or simple statistical associations is still a matter of 59 debate, most theories (see Patricia W Cheng (1997); Patricia W. Cheng and Buehner (2012), 60 for example) have accepted the ∆P metric as the basis of causal learning in humans; the ...
Preprint
Full-text available
Human experiments have demonstrated that instrumental performance of an action and the causal beliefs of the effectiveness of an action in producing a reward are correlated and controlled by the probability of an action leading to a reward. The animal literature, however, shows that instrumental performance under free-operant training differs even when these reward probabilities are matched while subjects undergo training under ratio or interval schedules of reward. In two experiments, we investigated whether causal beliefs would correlate with instrumental performance under interval and ratio schedules for matched reward probabilities. In both experiments we found that performance was higher under ratio than under interval training. However, causal beliefs were similar between these two conditions despite these differences in instrumental performance. When reward probabilities were increased by experimental manipulations in Experiment 2, the causal beliefs increased but performance decreased with respect to Experiment 1. This is evidence that instrumental performance and causal action-reward attribution may not follow from the same psychological process under free-operant training.
... This finding is consistent with work that shows that the perceived strength [28], salience [35], or exposure [16] of alternative causes, such as the context or other causes, competes with the cause under consideration for perceived control [15]. ...
Article
Full-text available
Background: Perceived control is strongly linked to healthy outcomes, mental healthiness, and psychological well-being. This is particularly important when people have little control over things that are happening to them. Perceived control studies have been performed extensively in laboratory settings and show that perceived control can be increased by experimental manipulations. Although these studies suggest that it may be possible to improve people's mental health by increasing their perceived control, there is very little evidence to date to suggest that perceived control can also be influenced in the real world. Objective: The first aim of this study was to test for evidence of a link between noncontrol situations and psychological well-being in the real world using a mobile phone app. The second and arguably more important aim of the study was to test whether a simple instructional intervention on the nature of alternative causes would enhance people's perceptions of their own control in these noncontrol situations. Methods: We implemented a behavioral action-outcome contingency judgment task using a mobile phone app. An opportunity sample of 106 healthy volunteers scoring low (n=56, no depression) or high (n=50, mild depression) on a depression scale participated. They were given no control over the occurrence of a low- or high-frequency stimulus that was embedded in everyday phone interactions during a typical day lasting 8 hours. The intervention involved instructions that either described a consistent alternative cause against which to assess their own control, or dynamic alternative causes of the outcome. Throughout the day, participants rated their own control over the stimulus using a quantitative judgment scale. Results: Participants with no evidence of depression overestimated their control, whereas those who were most depressed were more accurate in their control ratings. Instructions given to all participants about the nature of alternative causes significantly affected the pattern of perceived control ratings. Instructions describing discrete alternative causes enhanced perceived control for all participants, whereas dynamic alternative causes were linked to less perceived control. Conclusions: Perceptions of external causes are important to perceived control and can be used to enhance people's perceptions. Theoretically motivated interventions can be used to enhance perceived control using mobile phone apps. This is the first study to do so in a real-world setting.
... Indeed, there is a wealth of literature demonstrating the ability of humans to learn action-outcome contingencies (e.g. Allan & Jenkins, 1980;Dickinson, Shanks & Evenden, 1984). More recently, Elsner and Hommel (2004) demonstrated that participants who experienced a learning phase in which a tone followed a keypress with a high degree of contingency, showed faster reaction times, in a subsequent test phase, to execute that keypress in response to the tone relative to a keypress not previously associated with the tone. ...
Article
Full-text available
The sense of agency is defined as one's sense of control over one's actions and their consequences. A recent theory, the Control-Based Response Selection Framework (Karsh & Eitam, 2015a), suggests that actions associated with a high sense of agency are intrinsically rewarding and thus motivate response selection. Previous studies support this theory by demonstrating that factors impacting on sense of agency (e.g. probability of an outcome following an action) also motivate selection of actions. Here we report a novel test of the Control-Based Response Selection Framework in the domain of action-outcome contingency. The contingency between actions and their outcome has previously been demonstrated to impact the sense of agency, but its impact on the motivation to perform actions has not yet been examined. Participants were asked to press one of four buttons as randomly as possible. Each of the buttons was assigned a different probability of causing an outcome when pressed. Additionally, a contingency manipulation was employed where the probability of an outcome occurring in the absence of a button press was also varied in blocks throughout the experiment. Results demonstrated a significant influence of contingency on response speed, and a significant effect of probability on response selection, consistent with predictions from the Control-Based Response Selection Framework. Furthermore, some evidence was observed for a positive correlation between influence of contingency and autistic traits, with individuals with higher autistic traits showing a greater influence of contingency on reaction times. The current findings support the idea that actions associated with an increased sense of agency are intrinsically rewarding, and identify how individual differences may impact on this process.
... Another experiment by Wagner and Saavedra (Wagner 1969) replicated the result when the unconditioned stimulus was a food reward instead of shock. Both the sequential and simultaneous blocking effect have been found in humans (e.g., Dickinson, Shanks, & Evenden, 1984;Shanks, 1991). ...
... Second, there is parallel interest in alterations of associative learning that also accompany SCZ. Given that similar psychological learning mechanisms have been proposed to describe both causal learning and reward-based (reinforcement) learning (Dickinson, Shanks, & Evenden, 1984;Gershman, 2015), it may be that abnormalities in reward-based paradigms can reflect deficits in the former. ...
Article
Dysfunction of reward‐related neural circuitry in schizophrenia (SCZ) has been widely reported, and may provide insight into the motivational and cognitive disturbances that characterize the disorder. Although previous meta‐analyses of reward learning paradigms in SCZ have been performed, a meta‐analysis of whole‐brain coordinate maps in SCZ alone has not been conducted. In this study, we performed an activation likelihood estimate (ALE) meta‐analysis, and performed a follow‐up analysis of functional connectivity and functional decoding of identified regions. We report several salient findings that extend prior work in this area. First, an alteration in reward‐related activation was observed in the right ventral striatum, but this was not solely driven by hypoactivation in the SCZ group compared to healthy controls. Second, the region was characterized by functional connectivity primarily with the lateral prefrontal cortex and pre‐supplementary motor area (preSMA), as well as subcortical regions such as the thalamus which show structural deficits in SCZ. Finally, although the meta‐analysis showed no regions outside the ventral striatum to be significantly altered, regions with higher functional connectivity with the ventral striatum showed a greater number of subthreshold foci. Together, these findings confirm the alteration of ventral striatal function in SCZ, but suggest that a network‐based approach may assist future analysis of the functional underpinnings of the disorder.
... In a typical causal learning task, participants are first presented with a series of trials where certain stimuli are paired with certain outcomes, and these participants will then be asked to rate the likelihood of the stimuli predicting the outcomes (Luque and Vadillo, 2011). Under certain conditions, human causal learning is comparable to the underlying processes of associative learning in animal conditioning (Dickinson, Shanks and Evenden, 1984;Shanks and Dickinson, 1988; see Chapman and Robbins for review). ...
Article
Associative learning phenomena have been widely used to understand the deficits in selective attention in schizophrenia by using the personality trait, schizotypy, as a proxy. However, other personality traits such as anxiety and the Big 5 personality traits have been under-looked despite a comorbidity between schizophrenia/schizotypy and anxiety as well as the psychopathology links of the Big 5 traits. Moreover, there is evidence of different thinking styles exhibited by different cultures (e.g., individualistic and collectivistic cultures), where the majority of members in an individualistic culture learn and think in an analytical/elemental manner while the majority of members in a collectivistic culture have a predisposition to think and learn holistically/configurally. It is therefore proposed that other personality traits and cultural differences in thinking/learning can explain conflicting evidence found in the schizotypy and associative learning literature. Previous studies of variations in attention-driven associative learning have demonstrated an emphasis on latent inhibition and less so on blocking and learned predictiveness. Furthermore, there are very few studies that have attempted to reproduce two learning effects within the same individual. Therefore, Study 1 aimed to create a paradigm which can generate the effects of blocking and learned predictiveness within the same participant to first, fill the gap in the literature, and second, to develop a better, converging, understanding of the role of attention in learning. The results of this study found an effect of learned predictiveness but no effect of blocking. It was proposed that the effect of learned predictiveness somehow masked the effect of blocking, so Study 2 aimed to replicate the previous study but with only the blocking trials. The results still showed no blocking effect despite the removal of the learned predictiveness trials. There was a possibility that a within-compound association effect was the reason why blocking was not found. Therefore, in Study 3, the design of Study 2 was replicated but now with an addition of a Stage 3 where the blocked stimuli’s contingencies were switched, and if there really was a within-compound association, the ratings for these blocked stimuli would be reduced compared to the first test stage. The results showed at test stage 1, the ratings for the control stimuli were lower than the blocked stimuli, replicating results of Study 2. At test stage 2, with a change in contingency from Stage 3, the ratings for the blocked stimuli were reduced but it was still higher than the control stimuli, suggesting a within-compound association. Study 4 aimed to use a simpler blocking design (Kamin, 1969) to determine if the previous design used was too complicated. While there the ratings for the control stimuli were higher than the blocked stimuli, a paired samples t-test revealed no significance. The results from Studies 1-4 can be explained using acquired distinctiveness/acquired equivalence theories and the redundancy effect. Since attempts to demonstrate a blocking effect throughout Study 1 to 4 failed, Study 5 aimed to generate the effects of latent inhibition and learned predictiveness using a letter prediction task (Granger et al., 2016) and a food allergist task (Le Pelley and McLaren, 2003). Personality traits including the Big 5, schizotypy and anxiety were also measured. The participants were divided into individualistic and collectivistic groups using Hofstede’s database which sorts individuals via their nationalities. The results initially showed that participants, overall, exhibited both latent inhibition and learned predictiveness. When split by culture, only participants from the individualistic group showed both effects while participants from the collectivistic group showed only latent inhibition. There was also no correlation between latent inhibition and learned predictiveness overall and within groups. Moreover, there was an effect of impulsive nonconformity that was related to a greater magnitude of latent inhibition in the individualistic group. It was also found that participants high in conscientiousness from the individualistic group learned more about the relevant stimuli in the learned predictiveness task but participants from the collectivistic group learned about the relevant stimuli less. There was no evidence of anxiety predicting latent inhibition or learned predictiveness. Since Study 5 was exploratory by nature, Study 6 aimed to replicate the findings but specifically focussing on the relationship between personality traits and culture in latent inhibition. It could be seen from the results of this current study that latent inhibition was observed in both the individualistic and collectivistic group, replicating Study 5. This study was also conducted online due to Covid-19 and the results demonstrated that there was no difference in response times between the lab-based task and this online version. The effect of impulsive nonconformity predicting enhanced latent inhibition in participants from the individualistic group found in Study 5 was not replicated. Instead, it was revealed that conscientiousness predicted latent inhibition in the individualistic group and openness predicted latent inhibition in the collectivistic group. There was again no evidence of anxiety being related to the magnitude of latent inhibition. Study 7 was a replication of Study 5, but specifically investigating the relationship between the Big 5, schizotypy and cultural orientation and learned predictiveness. Since there were limitations to sorting participants into individualistic and collectivistic groups with their nationalities (Cohen, 2009; Kitayama and Uskul, 2011; Maisuwong, 2012), a cultural orientation questionnaire (Sharma, 2010) was employed to provide a more accurate measure of trait individualism and collectivism to provide further validity to the results found. The results showed a learned predictiveness effect in the individualistic group but not the collectivistic group, replicating results from Study 5. The effect of conscientiousness predicting the learning of relevant stimuli was not replicated but an effect of extraversion positively predicting the learning of irrelevant stimuli was found within the collectivistic group. There was also an effect of introvertive anhedonia negatively predicting the learning of irrelevant stimuli in the collectivistic group. The final chapter of this thesis begins with a summary of results from all the studies. It then discusses how the individual and cultural differences exhibited by participants influenced the results throughout Studies 1-7 and provides suggestions as to why evidence from previous literature was not always consistent. The strengths and limitations faced in this thesis are discussed and suggestions for future research described.
... The group with compound training received then extinction of the light, after which an increase in conditioned responding to the noise alone, even though they did not have any additional experience with the sound. RR of Pavlovian blocking, in which there is an increment in responding to the blocked cue after extinction of the blocking cue, has been reported both in human causal learning (Dickinson and Burke, 1996;Dickinson et al., 1984;Dopson et al., 2009;Shanks, 1985) and in fear conditioning (Blaisdell et al., 1999). In a similar manner, blocking of occasion setting may be retrospectively reevaluated if after blocking training, one element of the OS compound is manipulated to reduce its OS potential, that is, a blocked OS can be "unblocked", meaning that it can recover its capacity to disambiguate behavioral control by a CS. ...
Article
An occasion setter (OS) is a stimulus or context with the capacity to disambiguate an ambiguous conditioned stimulus (CS). Previous research has shown that OSs share some features with regular Pavlovian CSs. Amongst them, research has shown that OSs are subject to blocking; that is, a new OS exerts reduced behavioral control after training in compound with a previously established OS. Of additional interest, in Pavlovian blocking, it has been reported that a blocked CS comes to elicit conditioned responding after the extinction of the blocking CS. This is an example of retrospective revaluation, a family of phenomena in which the response to a specific stimulus is modified by training a related cue. Here, three experiments sought to extend the analogies between OS and Pavlovian conditioning by examining the blocking of OSs and its retrospective revaluation. In all experiments, an OS was established by pairing a CS with food in the presence of the OS, but not in its absence (i.e., positive OS). Blocking was then trained by presenting the OS in compound with a novel OS. Experiment 1 showed blocking of the second OS, but direct exposure to the blocking OS did not enhance responding to the second OS. Experiment 2 replicated the blocking effect but subsequent training of the blocking OS with a reversed contingency showed no retrospective revaluation. Experiment 3 examined whether blocking of the OS occurred with a novel CS during the compound phase. In this experiment blocking was again observed, but only when subjects were tested with the original CS. These results are discussed focusing on the underlying links at work in occasion setting.
... Blocking occurs when the previous learning of an association between a certain cue and an outcome interferes with the subsequent learning of an association between another cue and the same outcome. Specifically, in a typical blocking paradigm (e.g., Aitken, Larkin, & Dickinson, 2001;Dickinson, Hall, & Mackintosh, 1976;Dickinson, Shanks, & Evenden, 1984), subjects first learn to associate a certain cue (A) with a specific outcome (X). In a second phase, the cue is then presented alongside a second cue (B), both predicting the same outcome as before (X), rendering the second cue redundant. ...
Article
In the literature on learning, one of the most robust cue competition effects is blocking: the previous learning of a cue-outcome association prevents learning that other cues predict the same outcome if those cues are presented together with the first cue. In this research, we investigated blocking effects in mental state inference. Participants learned to diagnose the internal states of a target person based on the behaviors he displayed. Blocking effects were observed across several studies, such that, when participants had previously learned that a certain behavior predicted a certain internal state, they later failed to learn about the predictive value of other behaviors that were paired with the original behavioral cue. Implications are discussed for cognitive models of learning and cue competition, as well as for the social psychology of mental state inference.
... Les études de compétition entre indices typiques chez l'homme impliquent des objectifs d'apprentissage plutôt explicites (p. ex., Chapman & Robbins, 1990;Dickinson, Shanks, & Evenden, 1984;Gluck & Bower, 1988;Le Pelley & McLaren, 2001). Dans une expérience d'ombrage, les participants ont vu des mots et des formes composés en couleur (ombrage) ou seulement des mots colorés (mots-seulement) ou seulement des formes colorées (formes-seulement). ...
Article
Full-text available
Dans cet article, je passe en revue la littérature sur l'apprentissage incident des régularités simples de stimulus-réponse. L'article résume les travaux utilisant le paradigme d'apprentissage de contingence couleur-mot et les procédures connexes. Dans ce paradigme, un mot neutre coloré est présenté aux participants à chaque essai et ces derniers sont invités à ignorer le mot et à répondre à la couleur d'impression. Chaque mot distracteur est présenté le plus souvent dans une couleur cible (p. ex., « bouge » le plus souvent en bleu, etc.). L'apprentissage de ces contingences est indiqué par des réponses plus rapides et plus précises aux essais de forte contingence (dans lesquels le mot est présenté fréquemment avec sa couleur) par rapport aux essais de faible contingence. Cette procédure s'est avérée utile pour les recherches portant sur l'apprentissage incident. Le présent manuscrit résume le travail existant avec cette procédure d'apprentissage et celles connexes, et met en évidence les orientations émergentes. In this article, I review research on incidental learning of simple stimulus-response regularities. The article summarizes work with the colour-word contingency learning paradigm and related simple learning procedures. In the colour-word contingency learning paradigm participants are presented with a coloured neutral word on each trial and are asked to ignore the word and respond to the print colour (e.g., similar to a Stroop procedure). Distracting words are typically colour-unrelated neutral stimuli. However, each distracting word is presented most often in one target colour (e.g., "move" most often in blue, "sent" most often in green, etc.). Learning of these contingencies is indicated by faster and more accurate responses to high contingency trials (in which the word is presented with its frequent colour) relative to low contingency trials. This procedure has proven useful for investigations in incidental learning. The present manuscript summarizes the existing work with this (and related) learning procedures and highlights emerging directions.
... In the early 1980s, several authors suggested that the associative learning literature might provide a promising groundwork for understanding human causal learning (e.g., Alloy & Tabachnik, 1984;Dickinson et al., 1984;Shanks, 1985b;Shanks & Dickinson, 1987). In that vein, Shanks observed that "the laws of conditioning are fundamentally similar to the laws which determine causal relationships" (Shanks, 1985b, p. 106). ...
Article
The associative learning theory of Robert Rescorla and Allan Wagner has been duly celebrated for its 50-year reign as the predominant model in learning science. One special recognition is warranted: its close correspondence with David Hume's associative theory of causality judgment. Hume's rules by which causes come to suggest effects are not only embraced by the Rescorla-Wagner model, but their mechanistic account makes precise quantitative predictions that can be assessed by empirical evidence rather than by speculation and argumentation. Framed in this way, the Rescorla-Wagner model truly represents the scientific culmination of Hume's philosophical theory of causation. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
... Inhibitory learning occurs since the absence of an otherwise expected outcome (negative prediction error) results in a decrease in associative strengths for all cues present, resulting in cue B acquiring net negative associative strength as it decreases from zero. Although originally developed to explain conditioning in non-human animals, this account is also prominent in human causal learning (Dickinson, Shanks & Evenden, 1984). ...
Article
Full-text available
Inhibitory learning after feature negative training (A+/AB-) is typically measured by combining the Feature B with a separately trained excitor (e.g., C) in a summation test. Reduced responding to C is taken as evidence that B has properties directly opposite to those of C. However, in human causal learning, transfer of B's inhibitory properties to another excitor is modest and depends on individual differences in inferred causal structure. Here we ask whether instead of opposing processes, a summation test might instead be thought of in terms of generalization. Using an allergist task, we tested whether inhibitory transfer would be influenced by similarity. We found that transfer was greater when the test stimuli were from the same semantic category as the training stimuli (Experiments 1 and 2) and when the test excitor had previously been associated with the same outcome (Experiment 3). We also found that the similarity effect applied across all self-reported causal structures. We conclude it may be more helpful to consider transfer of inhibition as a form of conceptual generalization rather than the arithmetic summation of opposing processes. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
... Previous research [23][24][25][26][27][28] has shown that people are sensitive to contingency between events, and that contingency is used as a cue to make causal inferences; however, under some circumstances, people systematically deviate from the normative standard. Researchers have described two systematic deviations: the influence of the probability of effect occurrence [29][30][31][32][33][34], when the effect occurs frequently, the causal relationship tends to be overestimated ( Figure 1, panel B); and the influence of the probability of occurrence of the cause [24,35,36], when the probability of the cause is high, the contingency perceived between cause and effect is also high (Figure 1, panel C). ...
Article
Background The internet is a relevant source of health-related information. The huge amount of information available on the internet forces users to engage in an active process of information selection. Previous research conducted in the field of experimental psychology showed that information selection itself may promote the development of erroneous beliefs, even if the information collected does not. Objective The aim of this study was to assess the relationship between information searching strategy (ie, which cues are used to guide information retrieval) and causal inferences about health while controlling for the effect of additional information features. Methods We adapted a standard laboratory task that has previously been used in research on contingency learning to mimic an information searching situation. Participants (N=193) were asked to gather information to determine whether a fictitious drug caused an allergic reaction. They collected individual pieces of evidence in order to support or reject the causal relationship between the two events by inspecting individual cases in which the drug was or was not used or in which the allergic reaction appeared or not. Thus, one group (cause group, n=105) was allowed to sample information based on the potential cause, whereas a second group (effect group, n=88) was allowed to sample information based on the effect. Although participants could select which medical records they wanted to check—cases in which the medicine was used or not (in the cause group) or cases in which the effect appeared or not (in the effect group)—they all received similar evidence that indicated the absence of a causal link between the drug and the reaction. After observing 40 cases, they estimated the drug–allergic reaction causal relationship. Results Participants used different strategies for collecting information. In some cases, participants displayed a biased sampling strategy compatible with positive testing, that is, they required a high proportion of evidence in which the drug was administered (in the cause group) or in which the allergic reaction appeared (in the effect group). Biased strategies produced an overrepresentation of certain pieces of evidence at the detriment of the representation of others, which was associated with the accuracy of causal inferences. Thus, how the information was collected (sampling strategy) demonstrated a significant effect on causal inferences (F1,185=32.53, P<.001, η2p=0.15) suggesting that inferences of the causal relationship between events are related to how the information is gathered. Conclusions Mistaken beliefs about health may arise from accurate pieces of information partially because of the way in which information is collected. Patient or person autonomy in gathering health information through the internet, for instance, may contribute to the development of false beliefs from accurate pieces of information because search strategies can be biased.
... In an overshadowing design, two antecedents are presented before an outcome, and if the salience between them differs, subjects tend to attribute predictive value to the more salient event, at the expense of the less salient event (Mackintosh, 1976;Waldmann, 2001). Similarly, if an antecedent is trained in the presence of another antecedent that has previously been established as a predictor of the outcome, impaired learning about the new antecedent is observed, a phenomenon known as blocking (Dickinson et al., 1984;Kamin, 1968). A large history of research has confirmed the reliability of overshadowing and blocking across species and types of learning (including invertebrate species, e.g., Acebes et al., 2009, in garden snails;Prados et al., 2013, in planaria). ...
Article
Full-text available
Over the last 50 years, cue competition phenomena have shaped theoretical developments in animal and human learning. However, recent failures to observe competition effects in standard conditioning procedures, as well as the lengthy and ongoing debate surrounding cue competition in the spatial learning literature, have cast doubts on the generality of these phenomena. In the present study, we manipulated temporal contiguity between simultaneously trained predictors and outcomes (Experiments 1-4), and spatial contiguity between landmarks and goals in spatial learning (Supplemental Experiments 1 and 2; Experiment 5). Across different parametric variations, we observed overshadowing when temporal and spatial contiguity were strong, but no overshadowing when contiguity was weak. Thus, across temporal and spatial domains, we observed that contiguity is necessary for competition to occur, and that competition between cues presented simultaneously during learning is absent when these cues were either spatially or temporally discontiguous from the outcome. Consequently, we advance a model in which the contiguity between events is accounted for and which explains these results and reconciles the previously contradictory findings observed in spatial learning. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
... Process theories have been applied to understand the development of underlying control or contingency learning. For example, the Rescorla-Wagner model (RWM: Rescorla & Wagner, 1972) is an associative learning theory, which has been applied to contingency learning (e.g., Baker, Murphy, & Vallée-Tourangeau, 1996;Dickinson, Shanks, & Evenden, 1984;Vallée-Tourangeau et al., 1998) as well as learning of contingencies in animals. The RWM describes how associations develop between cues or actions and subsequent outcomes when they are paired, and how other stimuli (the context) also become associated with the same outcomes. ...
Article
Depression has been linked to weakened perceptions of control. The experimental evidence derives from tasks with exposure to stable action-outcome contingencies. One assumption has been that performance represents a general cognitive bias that might manifest itself by a global performance difference. Another view is that people have specific situational perceptions of control reflecting their recent actions and the contingencies to which they are currently experiencing. In an experiment with N = 179, participants acquired one of four action-outcome sequences (Constant or Variable). We measured how learning was reflected in ratings of control and probability of responding in relation to mood. In three experimental treatments, the overall contingency across training involved an average moderate degree of control (ΔP = 0.25), but differed in how control varied (Constant or one of two Variable treatments). A fourth, control treatment involved a Constant zero degree of control (ΔP = 0.00). Participants rated their control before, during and after each sequence, providing measures of pre-existing bias, ratings of control in specific situations and generalised control perceptions. Specific control ratings were only influenced by the contingency experience and not pre-existing bias. Higher scores on the Beck’s depression inventory were associated with weakened association between action and context ratings. Overall, these data suggest that human agency is related to rates of responding and that mood is related to a difference in sensitivity to the ratings of and responding to the context.
... The relationship between a cue and an outcome depends not only on their co-occurrence, but also the predictive qualities of other cues presented at the same time. According to associative accounts, selective learning effects arise due to simultaneously presented cues competing for a limited amount of associative strength with the outcome (Dickinson, Shanks & Evenden, 1984). ...
Article
People often fail to use base-rate information appropriately in decision-making. This is evident in the inverse base-rate effect, a phenomenon in which people tend to predict a rare outcome for a new and ambiguous combination of cues. While the effect was first reported in 1988, it has recently seen a renewed interest from researchers concerned with learning, attention and decision-making. However, some researchers have raised concerns that the effect arises in specific circumstances and is unlikely to provide insight into general learning and decision-making processes. In this review, we critically evaluate the evidence for and against the main explanations that have been proposed to explain the effect, and identify where this evidence is currently weak. We argue that concerns about the effect are not well supported by the data. Instead, the evidence supports the conclusion that the effect is a result of general mechanisms that provides a useful opportunity to understand the processes involved in learning and decision making. We discuss gaps in our knowledge and some promising avenues for future research, including the relevance of the effect to models of attentional change in learning, an area where the phenomenon promises to contribute new insights.
... Previous research has shown that, although people can use the contingency between cause and effect to infer causality [42][43][44][45][46][47] , under some circumstances, they can easily develop a causal illusion, that is, the belief that there is a causal connection between two events that are actually unrelated (i.e., non-contingent on each other). Causal illusions have been described as cognitive biases that appear in the general population and may underlie many relevant and societal problems, such as prejudice and pseudoscience 27,39,[48][49][50][51] . ...
Article
Full-text available
Previous research proposed that cognitive biases contribute to produce and maintain the symptoms exhibited by deluded patients. Specifically, the tendency to jump to conclusions (i.e., to stop collecting evidence soon before making a decision) has been claimed to contribute to delusion formation. Additionally, deluded patients show an abnormal understanding of cause-effect relationships, often leading to causal illusions (i.e., the belief that two events are causally connected, when they are not). Both types of bias appear in psychotic disorders, but also in healthy individuals. In two studies, we test the hypothesis that the two biases (jumping to conclusions and causal illusions) appear in the general population and correlate with each other. The rationale is based on current theories of associative learning that explain causal illusions as the result of a learning bias that tends to wear off as additional information is incorporated. We propose that participants with higher tendency to jump to conclusions will stop collecting information sooner in a causal learning study than those participants with lower tendency to jump to conclusions, which means that the former will not reach the learning asymptote, leading to biased judgments. The studies provide evidence in favour that the two biases are correlated but suggest that the proposed mechanism is not responsible for this association.
... 75 Credit assignment is widely regarded as a competitive process in which the best 76 predictor of the outcome acquires substantial credit over the course of learning at the 77 expense of other predictors (e.g., Honey Pearce, 1994;Miller & Matzel, 1988;Wagner, 1981;Pearce & Hall, 1980;Mackintosh, 80 1975; Rescorla & Wagner, 1972). In support of this notion is evidence that cues compete 81 for credit in a range of tasks (Kamin, 1968;Pavlov, 1927, Wagner et al., 1968, Rescorla, 82 1968, 1970) and species, from C. elegans to humans (Merritt et Beauchamp et al., 1991, Tobler et al., 2006Dickinson et al., 1984). However, it has also 85 . ...
Preprint
Full-text available
A fundamental assumption of learning theories is that the credit assigned to predictive cues is not simply determined by their probability of reinforcement, but by their ability to compete with other cues present during learning. This assumption has guided behavioral and neural science research for decades, and tremendous empirical and theoretical advances have been made identifying the mechanisms of cue competition. However, when learning conditions are not optimal (e.g., when training is massed), credit assignment is no longer competitive. This is a catastrophic failure of the learning system that exposes the individual's vulnerability to form spurious associations in the real world. Here, we uncover that cue competition can be rescued when conditions are suboptimal provided that the individual has agency over the learning experience. Our findings reveal a new connection between agency over learning and credit assignment to cues, and open new avenues of investigation into the underlying mechanisms.
... 1 (Evans, 2008;Evans & Stanovich, 2013) (e.g., Dickinson et al., 1984) (Shanks, 2007 for a review) Rescorla-Wagner (Rescorla & Wagner, 1972) (i.e., E1 C E2) ...
Article
Causal knowledge enables us to explain past events, to control present environment, and to predict future outcomes. Over the last decade, causal Bayes nets have been rec- ognized as a normative framework for causality and used as a psychological model to account for human causal learning and inference. This article provides an introduction to causal Bayes nets. According to causal Bayes nets, causal inference can be divided into three processes: (a) learning the structure of the causal network, (b) learning the strength of the causal relations, and (c) inferring the effect from the cause or the cause from the effect. For each process, I describe the predictions of causal Bayes nets, review experimental results, and suggest future directions. Although there are a few excep- tions (e.g., Markov violation), most of the results are consistent with the predictions of causal Bayes nets. The current problems of the Bayesian approach and its future perspective are discussed.
... There are a number of hypotheses to explain cue competition phenomena such as overshadowing and blocking (e.g., Mackintosh, 1975;Pearce & Hall, 1980;Rescorla & Wagner, 1972;Sutherland & Mackintosh, 1971), but the key point is that knowing the relationship between Stimulus A and the outcome hinders learning the relationship between Stimulus X and the outcome. Typical studies of cue competition in humans involve rather explicit learning objectives (e.g., Chapman & Robbins, 1990;Dickinson, Shanks, & Evenden, 1984;Gluck & Bower, 1988;Le Pelley & McLaren, 2001). Thus, Schmidt and De Houwer (2019b) used the colour-word contingency learning paradigm to examine whether overshadowing and blocking effects can be observed during incidental learning (i.e., where deliberate learning is not the goal of the task). ...
Article
Full-text available
In this article, I review research on incidental learning of simple stimulus-response regularities. The article summarizes work with the colour-word contingency learning paradigm and related simple learning procedures. In the colour-word contingency learning paradigm participants are presented with a coloured neutral word on each trial and are asked to ignore the word and respond to the print colour (e.g., similar to a Stroop procedure). Distracting words are typically colour-unrelated neutral stimuli. However, each distracting word is presented most often in one target colour (e.g., "move" most often in blue, "sent" most often in green, etc.). Learning of these contingencies is indicated by faster and more accurate responses to high contingency trials (in which the word is presented with its frequent colour) relative to low contingency trials. This procedure has proven useful for investigations in incidental learning. The present manuscript summarizes the existing work with this (and related) learning procedures and highlights emerging directions.
... Although developed to account for animal conditioning data, the Rescorla-Wagner model has also been suggested to provide a good account of causal learning in humans, where excitatory associative strength corresponds to the belief that the stimulus causes the outcome, and inhibitory associative strength corresponds to the belief that the stimulus prevents the outcome (Dickinson et al., 1984). We have recently been interested in exploring individual differences in the beliefs acquired by human participants in a causal learning task involving a feature negative (A+/ AB−) discrimination (Lee & Lovibond, 2021; see also Glautier & Brudan, 2019). ...
Article
Full-text available
We have previously reported that human participants trained with a simultaneous feature negative discrimination (intermixed A+ / AB- trials) show only modest transfer of inhibitory properties of the feature B to a separately trained excitor in a summation test (Lee & Lovibond, 2021). Self-reported causal structure suggested that many participants learned that the effect of the feature B was somewhat specific to the excitor it had been trained with (modulation), rather than learning that the feature prevented the outcome (prevention). This pattern is reminiscent of the distinction between negative occasion-setting and conditioned inhibition in the animal conditioning literature. However, in animals, occasion-setting is more commonly seen with a serial procedure in which the feature (B) precedes the training excitor (A). Accordingly, we ran three experiments to compare serial with simultaneous training in an allergist causal judgment task. Transfer in a summation test was stronger to a previously modulated test excitor compared to a simple excitor after both simultaneous and serial training. There was a numerical trend towards a larger effect in the serial group, but it failed to reach significance and the Bayes Factor indicated support for the null. Serial training had no differential effect on self-reported causal structure, and did not significantly reduce overall transfer. After both simultaneous and serial training, transfer was strongest in participants who reported a prevention structure, replicating and extending our previous results to a previously modulated excitor. These results suggest that serial feature negative training does not promote a qualitatively different inhibitory causal structure compared to simultaneous training in humans.
... In our vignette, participants received a scenario description and were subsequently asked to process, evaluate, and take action within the scenario. Through the differing scenario descriptions, we were able to activate participants' individual decision-making to capture the manipulated component's causal effects on subsequent judgments (Dickinson et al., 1984). More specifically, we assigned each participant to perform a strategic decision-making task, which was built on a validated decision scenario from Wood et al. (2017) and adapted to model the AI-augmented decision-making process. ...
Article
AI-augmented decision-making processes promise to transform strategic decisions around innovation management. However, despite a growing body of research on algorithmic management, very little is known about the behavioral effects of the AI-augmented decision-making process. This article utilizes a psychological perspective to research the interaction of artificial intelligence and human judgment, suggesting that AI-based advice affects human decision-making behavior and skews perceptions of decision outcomes. We present a vignette-based decision experiment involving 150 senior executives to examine the perception of AI-augmented decision-making at the individual level. In contrast to earlier research on algorithm aversion, we find that employing AI-based advisory systems positively affects choice behavior and amplifies decision quality perception. We further show how this overreliance on an AI-augmented decision-making process can be explained through both a higher degree of trust in the advisor and the attribution of a more structured process. This paper contributes to the emerging discussion as to the role of AI in management and the novel phenomenon of algorithm appreciation by investigating the interplay of human and artificial intelligence in strategic decision-making to show that AI-based advice is perceived as more trustworthy than human advice in an R&D investment context.
Thesis
Full-text available
Self-esteem and depression are strongly related, but the nature of this relation is still open to debate. The aim of this cumulative dissertation is to clarify the exact nature of the prospective relation between self-esteem characteristics (i.e., level, instability, and contingency) and depression by testing alternative, theoretically plausible models. In Article 1 (Sowislo & Orth, 2013) we investigated the prospective reciprocal relations of self-esteem level with depression and anxiety by means of a meta-analysis of longitudinal studies. We found consistent support for the vulnerability model of low self-esteem and depression (i.e., low self-esteem contributing to subsequent depression) but only weak support for the scar model of low self- esteem and depression (i.e., depression eroding self-esteem). Moderator analysis of the vulnerability effect of low self-esteem suggested its stability across different sample and study characteristics. Article 2 (Sowislo, Orth, & Meier, 2013) builds upon the results from Article 1 by testing whether other characteristics of self-esteem, namely self-esteem instability and contingency, have a vulnerability effect on depression over and above the vulnerability effect of self-esteem level. Two primary studies testing overarching models showed that only level of self-esteem predicted subsequent depression and that the characteristics of self-esteem did not interact in the prospective prediction of depression. Altogether, our findings suggest that low self-esteem, but not unstable and contingent self-esteem, acts as a stable and consistent vulnerability factor for depression. The dissertation contributes to resolving debates in personality psychology, refining theories of self-esteem and of depression and provides translational implications for clinical psychological interventions.
Article
Full-text available
A fundamental assumption of learning theories is that the credit assigned to predictive cues is not simply determined by their probability of reinforcement, but by their ability to compete with other cues present during learning. This assumption has guided behavioral and neural science research for decades, and tremendous empirical and theoretical advances have been made identifying the mechanisms of cue competition. However, when learning conditions are not optimal (e.g., when training is massed), cue competition is attenuated. This failure of the learning system exposes the individual’s vulnerability to form spurious associations in the real world. Here, we uncover that cue competition in rats can be rescued when conditions are suboptimal provided that the individual has agency over the learning experience. Our findings reveal a new effect of agency over learning on credit assignment among predictive cues, and open new avenues of investigation into the underlying mechanisms.
Article
Recent research has explored cue competition phenomena in social learning. In particular, blocking effects have been observed in the way people learn to infer someone's internal states from their behavioral cues: When people learn to associate a certain behavior with a certain internal state, this blocks their learning of subsequent behavioral cues that also predict the same state, when those cues are presented with the original behavior. In this research, we show that this blocking effect generalizes across targets, such that learning that a behavior predicts an internal state in a person hinders learning that other cues predict the same internal state in a different person, when both behaviors are presented simultaneously. This effect proved robust, and it was not moderated by the group membership of the targets.
Preprint
Full-text available
Humans possess a highly adaptive ability to draw inferences about the world by recognizing meaningful links between stimuli and events: making contingency judgements. We describe a systematic bias in contingency judgements that we label the negative contingency illusion in which individuals falsely judge a cue to be protective against an outcome. We demonstrate that the illusion arises when outcome probability is low and occurs when there is no actual relationship between cue and outcome and even when there is a modest positive relationship between cue and outcome. Such misjudgements may lead individuals to superstitious beliefs and could have major public health implications if they lead to the belief in and promotion of treatments that are ineffective or deleterious to the prevention and treatment of illness.
Article
Because humans live in a dynamic and evolving social world, modeling the factors that guide social behavior has remained a challenge for psychology. In contrast, much progress has been made on understanding some of the more basic elements of human behavior, such as associative learning and memory, which has been successfully modeled in other species. Here we argue that applying an associative learning approach to social behavior can offer valuable insights into the human moral experience. We propose that the basic principles of associative learning—conserved across a range of species—can, in many situations, help to explain seemingly complex human behaviors, including altruistic, cooperative, and selfish acts. We describe examples from the social decision-making literature using Pavlovian learning phenomena (e.g., extinction, cue competition, stimulus generalization) to detail how a history of positive or negative social outcomes influences cognitive and affective mechanisms that shape moral choice. Examining how we might understand social behaviors and their likely reliance on domain-general mechanisms can help to generate testable hypotheses to further understand how social value is learned, represented, and expressed behaviorally.
Article
In the current investigation we classified participants as inhibitors or non-inhibitors depending on the extent to which they showed conditioned inhibition in a context that had been used for extinction of a conditioned response. This classification enabled us to predict participant responses in a second experiment which used a different design and a different experimental task. In the second experiment a feature-negative discrimination survived reversal training of the feature to a greater extent in the non-inhibitors than in the inhibitors and this result was supported by Bayesian analyses. We propose that the fundamental distinction between inhibitors and non-inhibitors is based on a tendency to utilise first-order (direct associations) or second-order (occasion-setting) strategies when faced with ambiguous information and that this classification is a stable individual differences attribute.
Preprint
Full-text available
Several contemporary models of associative learning anticipate that the higher responding to a compound of two cues separately trained with a common outcome than to each of the cues alone -a summation effect-is modulated by the similarity between the cues forming the compound. Here, we explored this hypothesis in a series of causal learning experiments with humans. Participants were presented with two visual cues that separately predicted a common outcome and later asked for the outcome predicted by the compound of the two cues. Importantly, the cues’ similarity was varied between groups through changes in shape, spatial position, color, configuration and rotation. In variance with the predictions of these models, we observed similar and strong levels of summation in both groups across all manipulations of similarity (Experiments 1-5). The summation effect was significantly reduced by manipulations intended to impact assumptions about the causal independence of the cues forming the compound, but this reduction was independent of stimulus similarity (Experiment 6). These results are problematic for similarity-based models and can be more readily explained by rational approaches to causal learning.
Article
Rescorla and Wagner’s model of learning describes excitation and inhibition as symmetrical opposites. However, tasks used in human causal learning experiments, such as the allergist task, generally involve learning about cues leading to the presence or absence of the outcome, which may not reflect this assumption. This is important when considering learning effects which provide a challenge to this model, such as the redundancy effect. The redundancy effect describes higher causal ratings for the blocked cue X than for the uncorrelated cue Y in the design A+/AX+/BY+/CY-, the opposite pattern to that predicted by the Rescorla-Wagner model, which predicts higher associative strength for Y than for X. Crucially, this prediction depends on cue C gaining some inhibitory associative strength. In this manuscript, we used a task in which cues could have independent inhibitory effects on the outcome, to investigate whether a lack of inhibition was related to the redundancy effect. In Experiment 1, inhibition for C was not detected in the allergist task, supporting this possibility. Three further experiments using the alternative task showed that a lack of inhibition was related to the redundancy effect: the redundancy effect was smaller when C was rated as inhibitory. Individual variation in the strength of inhibition for C also determined the size of the redundancy effect. Given that weak inhibition was detected in the alternative scenario but not in the allergist task, we recommend carefully choosing the type of task used to investigate associative learning phenomena, as it may influence results.
Article
Full-text available
Over the last forty years, experimental support for different models of associative learning has come from a range of phenomena. Support for the Rescorla-Wagner (1972) model comes from blocking and overshadowing experiments; however, this model is unable to explain the findings of latent inhibition experiments. The Mackintosh (1975) model, on the other hand, is able to accommodate the findings from blocking, overshadowing and latent inhibition experiments, as well as discrimination learning, relative validity, learned irrelevance, intra-/extra-dimensional shift (IDS/EDS) and learned predictiveness experiments. The model proposed by Pearce and Hall (1980) is also able to explain the findings of blocking, overshadowing and latent inhibition experiments, but in addition to this it is also able to accommodate the effects of partial reinforcement and negative transfer. In an attempt to unify the theories into a single model that is able to explain all the aforementioned phenomena, Le Pelley (2004) proposed a hybrid model of associative learning, but it was not easily able to incorporate the effects of learned value. Alternatively, Esber and Haselgrove (2011) proposed a model that reconciles the influence of predictiveness and uncertainty into a single mechanism for attentional allocation, and this model was better able to explain the experimental findings of learned value. Theories of associative learning claim that a cue’s predictive validity determines the amount of attention it attracts and to what extent it is subsequently learned about (e.g. Mackintosh, 1975; Pearce & Hall, 1980). In Chapter 2, using eye-tracking methodology during a learned predictiveness task, several measures of overt attention were recorded and compared on trials where the predictive contingency was certain or less certain. Findings revealed that, at a within-trial level, good predictors of an outcome attracted more attention compared to irrelevant cues. Although, at a between-trial level, uncertain trials attracted more attention compared to certain trials. These findings provide support for the conflicting attentional modulation predictions made by the Mackintosh (1975) and Pearce-Hall (1980) models. Consequently, these findings can only be fully explained by appealing to a model of associative learning that incorporates both the principles of predictiveness and uncertainty (e.g. Le Pelley, 2004; Esber & Haselgrove, 2011). Prior to eye-tracking becoming more widely available as a measure of overt visual attention, stimulus associability was used as an indirect measure of attention since it is assumed that the speed at which a stimulus is learned about reflects the amount of attention it attracts. This is demonstrated in the IDS/EDS task which consistently finds that IDS are easier than EDS because in the IDS condition the higher associability of the predictive dimension in Stage 1 facilitates learning when generalised into Stage 2. Until now, eye gaze during an IDS/EDS task has not been investigated to determine whether the effect results from a shift in overt attention from Stage 1 into Stage 2. Chapter 3 revealed that participants acquired an attentional bias towards predictive cues in Stage 1 which transferred into Stage 2; however, in the EDS condition this bias was maintained only very briefly. Eye-tracking during learned predictiveness tasks using adult participants has revealed that cues which are good predictors of an outcome attract more overt visual attention than cues which are irrelevant. However, thus far, little research has investigated whether good predictors of reinforcement and non-reinforcement show a comparable effect. Moreover, it is currently unclear whether children and non-human animals demonstrate the learned predictiveness effect. Chapter 4 employed the same design and stimuli to examine eye gaze towards cues during a simple learned predictiveness task (AX+, AY+, BX-, BY-) in adults, children and an orangutan. Results revealed that all participants demonstrated the learned predictiveness effect, directing more attention towards cues that were good predictors of the outcome compared with cues that were irrelevant. However, for adult humans this effect was only present on reinforced trials and questionnaire data suggested they had only learned about one of the predictive contingencies. Contemporary discussions of associative learning have emphasised the importance of a cue’s predictive relevance in determining learned variations in attention. However, most theoretical accounts of the effect do not capture the notion of prediction – only associative strength, or relative associative strength (e.g. Mackintosh, 1975). In Chapter 5, letters were established as congruent or incongruent cues of other letters presented simultaneously or serially with a target cue. Results revealed no difference in the amount of attention directed towards congruent and incongruent cues if stimuli were presented simultaneously or serially when participants were required to respond to the identity of the target cue. However, an attentional bias towards congruent cues compared to incongruent cues was found when cues were presented serially, if participants were permitted to predict the identity of the target before its onset.
Article
Full-text available
Human experiments have demonstrated that instrumental performance of an action and the causal beliefs of the effectiveness of an action in producing a reward are correlated and controlled by the probability of an action leading to a reward. The animal literature, however, shows that instrumental performance under free-operant training differs even when the reward probabilities are matched while subjects undergo training under ratio or interval schedules of reward. In two experiments, we investigated whether causal beliefs would correlate with instrumental performance under ratio and interval schedules for matched reward probabilities. In both experiments, we found that performance was higher under ratio than under interval training. However, causal beliefs were similar between these two conditions despite these differences in instrumental performance. When reward probabilities were increased by experimental manipulations in Experiment 2, the causal beliefs increased but performance decreased with respect to Experiment 1. This is evidence that instrumental performance and causal action-reward attribution may not follow from the same psychological process under free-operant training.
Article
Full-text available
Review of the literature indicates that, according to theories of selective attention, learning about a stimulus depends on attending to that stimulus; this is represented in 2-stage models by saying that Ss switch in analyzers as well as learning stimulus-response associations. It is argued that this assumption, however, is equally well represented in a formal model by the incorporation of a stimulus-specific learning-rate parameter, a, into the equations describing changes in the associative strength of stimuli. Previous theories of selective attention have also assumed that (a) Ss learn to attend to and ignore relevant and irrelevant stimuli (i.e., that a may increase or decrease depending on the correlation of a stimulus with reinforcement); and (b) there is an inverse relationship between the probabilities of attending to different stimuli (i.e., that an increase in a to one stimulus is accompanied by a decrease in a to others). The first assumption has been used to explain the phenomena of acquired distinctiveness and dimensional transfer, the second to explain those of overshadowing and blocking. It is argued that although the first assumption is justified by the data, the second is not: Overshadowing and blocking are better explained by the choice of an appropriate rule for changing a, such that a decreases to stimuli that signal no change from the probability of reinforcement predicted by other stimuli. (65 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Male albino rats were exposed to negative correlations of tone CSs and shock USs. When the number of CSs and unpaired USs was held constant, the ability of the CS to resist subsequent excitatory conditioning declined as a function of the number or proportion of CSs paired with USs; so, too, did the ability of the CS to disrupt excitation to a second CS. In one treatment, in which the rate of USs in CS presence approached that in CS absence, the CS significantly enhanced excitation to a second CS. It is concluded that the rate of USs in CS presence need not be zero for a negative correlation to produce inhibitory effects. Also discussed is the possibility that positive and negative correlation procedures may have asymmetrical effects, at least preasymptotically.
Chapter
Instead of specifying the notion of information formally, this chapter examines some of the empirical operations and results which have led investigators to the intuition that information matters in conditioning. The chapter then suggests a theory, which acknowledges the importance of these operations in producing conditioning and discuss some empirical assessments of that theory. To the degree that the theory accounts for those operations which suggest the intuition of information, it can be viewed as an explication of that intuition, however, that notion does not enter into the theory itself. As the theory is explored, a number of points are discovered at which it provides a better account of the data than does the original intuition. Consequently, it is suggested that although the informational intuition serves an initially useful purpose, it does not provide an adequate conceptualization around which the understanding of Pavlovian conditioning can be organized.
Article
Presents a summary and synthesis of the author's work on attribution theory concerning the mechanisms involved in the process of causal explanations. The attribution theory is related to studies of social perception, self-perception, and psychological epistemology. Two systematic statements of attribution theory are described, discussed, and illustrated with empirical data: the covariation and the configuration concepts. Some problems for attribution theory are considered, including the interplay between preconceptions and new information, simple vs. complex schemata, attribution of covariation among causes, and illusions in attributions. The role of attribution in decision making and behavior is discussed. (56 ref.) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Previous investigations have shown that perceived contingency between responses and outcomes increases with the frequency of positive outcomes in the absence of actual contingency. In the present experiment with 30 undergraduates, judgments were obtained when both response alternatives were active choices (as in previous experiments), and when one alternative was to make no response. In the latter case, judgments were more accurate and less influenced by the frequency of positive outcomes. This result was predicted by the hypothesis that when both alternatives are active, Ss tend to assume that in the absence of any response no positive outcomes would occur. This faulty assumption, which is hypothesized to be one source of distortion in the judgment of contingency, is ruled out when no response is an explicit alternative within the task. (French abstract) (9 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
this essay develops the idea that attributions and causal inferences, particularly those made on the basis of partial information, are derived from causal schemata after specifying the general properties of causal schemata, this paper reviews the relevant social psychological literature for what it has to say about the types of schemata likely to be found in the typical attributor's repertoire delineate these potentialities of a causal schematic analysis of the attribution process as well as to indicate some of its major limitations (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Discusses 2 explanations of belongingness. One explanation postulates the evolution of event-specific associative mechanisms which associate events related in the environment. Another explanation suggests that certain events covary in a manner which improves associability. 2 forms of event-covariance are considered: similarity in the location and in temporal intensity patterns of the to-be-associated events. Published data are reviewed and new data are presented which indicate that these factors facilitate associability. Distinctions are drawn between event-specific associative mechanisms (activated by the to-be-associated events) and nonspecific associative mechanisms (activated by forms of event covariance). It is suggested that nonspecific mechanisms evolved to handle learning about real-world causal relationships which are variable with respect to the events involved. The explanatory power of this approach is applied to the acquisition of avoidance responses. (46 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Two experiments investigated 240 undergraduates' competence at judging event covariations under conditions approximating the memory demands found in everyday judgment settings. Ss were presented with sequences of individual event/state combinations and were asked either to record the event frequencies as they were presented (Exp I), to estimate the frequency information from memory (Exps I and II), or to estimate frequencies of event combinations and distractor cues from memory (Exp II). Comparisons of covariation judgment in these conditions showed that judgment accuracy deteriorated as memory demands increased. Further analyses suggested that the memory effect was due to poor recall of the frequency of event/state combinations and to a tendency to shift to simpler but less accurate judgment strategies. It is concluded that identification of event relationships in everyday settings is likely to be an error-ridden process. (28 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Examines the varied measures of contingency that have appeared in the psychological judgment literature concerning binary variables. It is argued that accurate judgments about related variables should not be used to infer that the judgments are based on appropriate information. (9 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Three experiments with 224 undergraduates investigated the sources of evidence people use in making inferences about causality in complex and uncertain situations. Ss were asked to evaluate the strength of the causal relationship between an outcome and some hypothesized cause of that outcome, on the basis of a given body of evidence. The evidence consisted of descriptions of a number of situations in which the outcome had occurred and had failed to occur and, for each situation, information about the presence and absence of some of the events that might have affected that outcome. Over problems and over experiments, the nature and strength of evidence supporting the causal role of the hypothesized cause varied along many dimensions. Using regression modeling, 5 evidence types that together gave a good account of Ss' judgments were found. Four types directly concerned the relation between the hypothesized cause and the outcome, and the 5th type reflected a secondary source of information about the hypothesized cause. This single linear model accounted for 84–90% of the variance in each problem set. Thus, although Ss were far more rational than they themselves claimed to have been, they were not immune to common biases found in other tasks. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
The function of the hippocampus in conditioning is portrayed in terms of an extension of N. J. Mackintosh's (see record 1975-26802-001) attention theory, which describes the evolution of the salience (associability) of each stimulus in the situation, including the context, and its predictive associative relationship to itself and all other stimuli. In terms of the model, the hippocampus is essential for computations that reduce salience when a stimulus is presented in the context of other stimuli that are better predictors of events. The model is applied to the phenomena of latent inhibition and blocking. (15 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
In the first of two experiments on the concept of correlation in adult subjects, the subjects' frequency estimates and inferences of relationship were studied relative to five different 2 × 2 distributions, each presented in a fixed sequence. In experiment II, the subjects' spontaneous strategies in subdividing and analyzing one 2 × 2 distribution were studied in a free situation. It is concluded that adult subjects with no statistical training apparently have no adequate concept of correlation (based on the ratio of the two pairs of diagonal frequencies), and that, in so far as they reason statistically at all, they tend to depend exclusively on the frequency of ++ cases in judging relationship. The need for studies involving ordinal scale and fully quantified variates is stressed.
Article
Studied the effects of preexposure to truly random (tr) and to traditional single-stimulus control treatments on subsequent acquisition of an estes-skinner cer in 132 male hooded rats. When a high-density tr schedule was employed, the tr cs appeared to acquire excitatory properties, which facilitated cer acquisition; other control treatments tended to retard subsequent cer acquisition. When a low-density tr schedule was employed, the tr cs retarded subsequent cer acquisition to a greater degree than did prior exposure to either cs or ucs alone. The low-density tr schedule, unlike an explicitly unpaired treatment, did not produce a cs with conditioned inhibitory properties. Results indicate that properties of the cs used in a tr control procedure depend upon variables associated with stimulus density.
Article
2 EXPERIMENTS WITH MALE SPRAGUE-DAWLEY RATS INDICATE THAT CS-UCS CONTINGENCY IS AN IMPORTANT DETERMINANT OF FEAR CONDITIONING AND THAT PRESENTATION OF UCS IN THE ABSENCE OF CS INTERFERES WITH FEAR CONDITIONING. IN EXP. I, EQUAL PROBABILITY OF A SHOCK UCS IN THE PRESENCE AND ABSENCE OF A TONE CS PRODUCED NO CONDITIONED EMOTIONAL RESPONSE SUPPRESSION TO CS; THE SAME PROBABILITY OF UCS GIVEN ONLY DURING CS PRODUCED SUBSTANTIAL CONDITIONING. IN EXP. II, WHICH EXPLORED 4 DIFFERENT PROBABILITIES OF UCS IN THE PRESENCE AND ABSENCE OF CS, AMOUNT OF CONDITIONING WAS HIGHER THE GREATER THE PROBABILITY OF UCS DURING CS AND WAS LOWER THE GREATER THE PROBABILITY OF UCS IN THE ABSENCE OF CS; WHEN THE 2 PROBABILITIES WERE EQUAL, NO CONDITIONING RESULTED.
Article
Reports 2 experiments which indicate that negative contingencies between CSs and shock set up conditioned inhibitors. In Exp. I with 48 male Sprague-Dawley rats, this inhibition was measured by retardation in the subsequent acquisition of a conditioned emotional response (CER) to the CS. Stimuli with greater negative CS-UCS contingencies were more retarded in CER acquisition; various control procedures were employed. In Exp II with 32 Sprague-Dawley rats, inhibition was measured by a summation technique. CSs with a history of greater negative relations to shock were more disruptive of the CER normally elicited by a 2nd CS. Taken together, the experiments support the general hypothesis that CS-UCS contingency is an important factor in fear conditioning.
Article
Previous experiments have shown that educated adults generally fail to show an intuitive appreciation of correlation or contingency when judging the relation between events on the basis of a serial presentation. The effect on judgment of displaying information serially or in a summary form was examined. In contrast with some previous experiments, the events to be judged were identified in a way which should strongly suggest that the operation of chance must be taken into account. The Ss judged the amount of control exerted by cloud seeding over rainfall. The events (seeding or no seeding followed by rain or no rain) were presented to 1 group only serially, to a 2nd group in both ways with the serial display preceding the summary. Only in the group which received the summary without the serial display were the judgments of a majority of Ss more consistent with an appropriate rule of judgment involving a comparison of probabilities than with 1 or another of several inappropriate rules involving the frequency of certain favourable events.
Article
Part 1 of this discussion summarizes several formal models of exicitatory classical conditioning. It is suggested that a central problem for all of them is the explanation of cases in which learning does not occur in spite of the fact that the CS is a signal for the reinforcer. A new model is proposed that deals with this problem by specifying that certain procedures cause a CS to lose effectiveness; in particular, it is argued that a CS will lose associability when its consequences are accurately predicted. In contrast to other current models, the effectiveness of the reinforcer remains constant throughout conditioning. Part 2 presents a reformulation of the nature of the learning produced by inhibitory-conditioning procedures and a discussion of the way in which such learning can be accommodated within the model outlined for excitatory learning. (47 ref)
Quantitative Studies in Operant Behavior: Acquisition
  • A R Commons
  • R J Wagner
  • Herrnstein
Commons, A. R. Wagner and R. J. Herrnstein (Eds.), Quantitative Studies in Operant Behavior: Acquisition. Cambridge, Mass.: Ballinger.
Selective association and conditioning
  • L J Kamin
Kamin, L. J. (1969). Selective association and conditioning. In N. J. Mackintosh and W. K. Honig (Eds.), Fundamental Issues in Associative Learning. Halifax: Dalhousie University Press.
  • Moore J. W.