ArticlePDF Available

Abstract and Figures

Recent evidence suggests that if a deterministic description of the events leading up to a morally questionable action is couched in mechanistic, reductionistic, concrete and/or emotionally salient terms, people are more inclined toward compatibilism than when those descriptions use non-mechanistic, non-reductionistic, abstract and/or emotionally neutral terms. To explain these results, it has been suggested that descriptions of the first kind are processed by a concrete cognitive system, while those of the second kind are processed by an abstract cognitive system. The current paper reports the results of three studies exploring whether or not considerations about possible future consequences of holding an agent responsible at a present time affect people’s judgments of responsibility. The results obtained suggest first that the concrete system does not produce compatibilist judgments of responsibility unconditionally, even when facing appropriately mechanistic, reductionistic, emotionally loaded and concretely worded deterministic scenarios. Second, these results suggest that considerations about possible future consequences for innocent third parties that may follow as a result of holding an agent responsible affect people’s judgment as to whether or not the agent is responsible for what she did. Finally, it is proposed that these results compliment extant evidence on the so-called “Side-effect effect”, as they suggest that emotional reactions toward possible future side effects influence people’s judgment of responsibility. The impact of these results for philosophy and moral psychology is discussed.
Content may be subject to copyright.
The Effect of What We Think may Happen
on our Judgments of Responsibility
Felipe De Brigard &William J. Brady
Published online: 6 February 2013
#Springer Science+Business Media Dordrecht 2013
Abstract Recent evidence suggests that if a deterministic description of the events
leading up to a morally questionable action is couched in mechanistic, reductionistic,
concrete and/or emotionally salient terms, people are more inclined toward compatibil-
ism than when those descriptions use non-mechanistic, non-reductionistic, abstract
and/or emotionally neutral terms. To explain these results, it has been suggested that
descriptions of the first kind are processed by a concrete cognitive system, while those of
the second kind are processed by an abstract cognitive system. The current paper reports
the results of three studies exploring whether or not considerations about possible future
consequences of holding an agent responsible at a present time affect peoplesjudg-
ments of responsibility. The results obtained suggest first that the concrete system does
not produce compatibilist judgments of responsibility unconditionally, even when
facing appropriately mechanistic, reductionistic, emotionally loaded and concretely
worded deterministic scenarios. Second, these results suggest that considerations about
possible future consequences for innocent third parties that may follow as a result of
holding an agent responsible affect peoples judgment as to whether or not the agent is
responsible for what she did. Finally, it is proposed that these results compliment extant
evidence on the so-called Side-effect effect, as they suggest that emotional reactions
toward possible future side effects influence peoples judgment of responsibility. The
impact of these results for philosophy and moral psychology is discussed.
1 Introduction
Philosophical and psychological research on free will and moral responsibility has
increased rapidly (Woolfolk et al. 2006; Nichols 2011). One such prominent area of
research pertains to peoples judgments of responsibility when considering
Rev.Phil.Psych. (2013) 4:259269
DOI 10.1007/s13164-013-0133-8
F. De Brigard (*)
Department of Psychology, Harvard University, 860 William James Hall 33 Kirkland St.,
Cambridge, MA 02139, USA
e-mail: brigard@wjh.harvard.edu
W. J. Brady
Department of Psychology, New York University, New York, NY, USA
deterministic versus non-deterministic scenarios (Nahmias et al. 2007). According to
many philosophers (e.g., Kane 1996; Pereboom 2001), when confronted with the
possibility of determinismroughly, the view according to which every event is
entailed by past events and the laws of naturepeoples intuitive reaction is to side
with incompatibilism about moral-responsibility: the view according to which deter-
minism and moral responsibility are incompatible (Fischer 1994). The thought is that,
by and large, peoples default belief is that an agent should not be held responsible for
an action she was bound to do. However, results from recent studies cast doubt upon
this claim. For example, Nahmias and collaborators (Nahmias et al. 2005; Nahmias et
al. 2006) asked participants to make judgments of moral responsibility following the
presentation of deterministic scenarios, and they found that, contrary to the received
philosophical view, peoples judgments were actually in line with compatibilism.
Nonetheless, these results alone do not establish that people are naturally compa-
tibilists. In fact, further studies have shown that if events that preceded the action in a
given case are couched in concrete and emotionally salient terms, people tend to
produce assessments in accordance with compatibilism. However, when scenarios are
abstract and emotionally void peoples intuitions tend to be incompatibilist (Nichols
and Knobe 2007; Sarkissian et al. 2010). Similarly, when the events that preceded the
action are described as occurring in a possible but non-actual world, peoples
intuitions are more incompatibilist than when events preceding the action are de-
scribed as occurring in the actual world (Roskies and Nichols 2008). Complementary
results have been reported in studies assessing peoples judgments of responsibility
when presented with reductionistic versus non-reductionistic scenarios (Nahmias et
al. 2007). These results suggest that when people face cases in which the events
preceding the action are described in purely neurological and non-psychological
terms, peoples assessments of responsibility vary as a function of how concrete
and/or emotionally salient the case is described. When the action in question is
described in purely abstract terms, peoples responses accord with incompatibilism,
whereas if the action is described in concrete and emotionally charged terms, peoples
responses tend toward compatibilism (De Brigard et al. 2009).
Taken together, the results of these and similar studies suggest that if the deter-
ministic description of the events leading up to a morally questionable action is
couched in mechanistic, reductionistic, concrete and emotionally salient terms, peo-
ple are more likely to hold the agent responsible than when those descriptions use
non-mechanistic, non-reductionistic, abstract and emotionally neutral terms. For
some theorists, these results are indicative of a more basic psychological asymmetry
that cuts across the incompatibilism / compatibilism chiasm. This asymmetric pattern
of responses is said to be the result of two putatively different underlying cognitive
systems driving our judgments of moral responsibility (e.g., Nichols and Knobe
2007; Sinnott-Armstrong 2008). One such systemcall it the concrete system
would be sensitive to producing judgments of moral responsibility for mechanistic,
reductionistic, concrete, and emotionally loaded deterministic scenarios, whereas the
otherthe abstract systemwould be sensitive to producing judgments of moral
responsibility for general, abstract and emotionally void deterministic descriptions.
However, several questions about the precise behavior of such systems remain
unanswered. One such question is whether the concrete system produces compatibi-
list judgments of responsibility in every mechanistic, reductionistic, emotionally
260 F. Brigard, W.J. Brady
loaded and concrete deterministic (MRECD) description unconditionally, or whether
there are conditions under which MRECD descriptions can nonetheless lead to
incompatibilist judgments of responsibility. The studies we present here attempt to
shed light on this issue by examining whether considerations about possible future
consequences of holding an agent responsible
1
may affect peoples judgments of
responsibility.
This particular manipulation was chosen for three reasons. First, since the
question we wanted to explore was whether MRECD descriptions of the causal
history leading up to an action unconditionally generate compatibilist judgments
of responsibility, we needed a way to keep the MRECD description fixed
between conditions while still looking for differences in judgments. We were
able to do so by manipulating considerations about possible future events.
Second, the question of whether possible future events that may occur as a
result of holding an agent responsible affect peoples judgment of responsibility
remains largely unexplored. In fact, almost all studies on free will and moral
responsibility have focused exclusively on the way in which peoples assessment
of the causal history of events preceding the to-be-evaluated action plays a role
in their judgments of responsibility (Nichols 2011). Specifically, researchers have
been interested in discerning which features of the description of the causal
history preceding an agents action affect peoples judgment of his or her
responsibility. A cursory look at the vignettes employed in studies such
Nahmias et al. (2005,2006) and Nichols and Knobe (2007), to name a few,
reveals that only events preceding the action upon which the protagonist is
judged are ever mentioned, and that there is no reference about possible future
events. But there is no reason to think that ordinary people only care about the
events immediately preceding the agents action, other than philosophersinsis-
tence that possible consequences that might ensue as a result of holding or
failing to hold an agent responsible are irrelevant when it comes to judgments
of responsibility. However, it is unclear whether ordinary people only keep in
mind the considerations philosophers assume they ought to keep in mind when
making judgments of responsibility. As such, the experiments reported here are
designed to explore the effect of considering possible future consequences of
holding an agent responsible on peoples judgment of responsibility.
Finally, a third motivation for choosing this particular manipulation comes from
recent evidence suggesting that moral considerations on the side effects of an agents
action affect peoples judgments on whether or not such actions were done intention-
ally (Knobe 2003), whether or not agents performed such actions knowingly (Beebe
and Buckwalter 2010), and even whether or not the agent was responsible for such
actions (Björnsoon and Persson In press). However, in virtually all the studies on this
so-called side-effect effector Knobe effect, (Feltz 2007) such side effects are
always part of the history that preceded the action upon which the agents responsi-
bility is evaluated. Whether or not considerations of possible future side effects also
affect judgments of responsibility remains an open question. The studies we present
here attempt to shed light on this issue as well.
1
To avoid excessive verbiage throughout the paper, whenever we refer to the consequences of holding an
agent responsible, we mean holding an agent responsible at the present time.
The Effect of What We Think 261
2 Study 1
2.1 Procedure
Fifty-four undergraduates from UNC-Chapel Hill (Age M= 20.8, SD = 4.1) voluntarily
participated in this study. None of the participants were philosophy or psychology
majors. All participants read a vignette (see below) describing a situation in which an
agent performed a morally questionable action as a result of a deterministic chain of
events. The description was couched in MRECD terms, akin to those used in many of
the studies described above (e.g., Nahmias et al. 2007;Nichols2011). Half of the
participants were randomly assigned to the Better condition, in which the vignette stated
that the status of an innocent third party may improve if the agent is found responsible.
The other half were assigned to the Worse condition, were it was stated that the status of
an innocent third party may worsen if the agent is found responsible. The specific
vignette used was the following (Better condition in parenthesis and italics):
Mary is the single mother of two: Mark, 7, Sally, 4. Mary works most of the
day, and although she is known for being fairly patient and good natured, over
the last year she has exhibited some unusually aggressive behavior toward her
neighbor. Last week, when she came back from work late at night, she couldnt
drive into her garage because her neighbor had blocked her driveway with his
new BMW. Enraged, she stepped on the gas pedal and crashed her car into her
neighbors. Unfortunately, her neighbor was still inside the car (it was too dark
for anyone to see him), and both his legs were seriously broken in several
places. Now he is not only suing her for several thousand dollars, but hes also
pressing charges. However, a neurologist examined her brain and discovered
that, in the last year, Mary has been developing a rare tumor in her frontal lobe.
Since the frontal lobe is necessary for emotional suppressionthat is, the
capacity to control onesemotionsthe neurologist claims that, unlike a
healthy person, Mary was completely unable to control her rage and her desire
to smash the car. In fact, he says, any person with this kind of tumor, facing
the exact same situation, would have done exactly what Mary did. She couldnt
have done otherwise. If Mary is found responsible for her actions, she may be
sent to a federal medical facility for the next 6 months. There she could receive
medical treatment, but she wont be able to see her children. Unfortunately
[Fortunately], during that time, they would be living with Social Services [Aunt
Elizabeth], in what might be a much worse [better] environment for them.
Participants were asked to say whether they agree or disagree with the following
statement: Mary is morally responsible for crashing her car into her neighbors.A
Lickert scale running from 1 to 7 was used to record participantsanswers, with 1
corresponding to Disagree,7toAgreeand 4 to Undecided.
2.2 Results and Discussion
Participants were more likely to say that Mary was responsible for running her
car into her neighborsintheBetter condition (M=5.30; SD=1.2) than they
262 F. Brigard, W.J. Brady
were in the Wors e condition (M= 3.15; SD =1.7), t(26)=9.605, p<.001, r = .59
(Fig. 1a). These results suggest that, when facing the possibility of an innocent
third partyslifeimproving as a result of holding the agent responsible, most
participants tended to hold the agent responsible even when it was clear that
her decision could be traced back to factorslike her tumorwhich were
outside of her control (compatibilist judgment). However, when the third party
could be worse off as a result of holding the agent responsible, most people
tended to not hold her responsible (incompatibilist judgment). These results
suggest that the alleged concrete system, which supposedly elicits compatibilist
judgments of responsibility for MRECD descriptions, does not operate uncon-
ditionally. Considerations of possible future consequences can sway such judg-
ments toward incompatibilism.
Moreover, these results also suggest that factors other than the causal history
leading up the agents action can influence peoples assessment of an agents respon-
sibility. In particular, these results suggest that the possible consequences that may
follow as a result of holding a person responsible can be factored into peoples
judgments of responsibility. Finally, these results also suggest that moral consider-
ations on side effects need not be confined to side effects that have already occurred,
as possible future consequences that have not yet occurred can also sway peoples
judgment of responsibility in a manner that is consistent with extant results on the
side-effect effect (Knobe et al. 2011).
A potential difficulty raised by the vignette employed in Study 1 is that it
featured an agent with a neural pathology. Previous studies have shown that the
use of cases featuring agents with neural pathologies yields mixed results (e.g.,
Greene and Cohen 2004; De Brigard et al. 2009). To address this worry, a second
study was conducted.
Fig. 1 Mean responses for both Better and Worse conditions. adisplays mean responses for Study 1.b
displays mean responses for Study 2.cdisplays mean responses for Study 3. Error bars depict standard
errors of the mean
The Effect of What We Think 263
3 Study 2
3.1 Procedure
Fifty undergraduates from UNC-Chapel Hill (Age M= 21.7, SD = 5.2) participated in
this study. The methodology was identical to that in Study 1, except that this time
around the vignettes depicted scenarios that did not rely on a pathological disorder.
Instead, the MRECD description of the events leading up to the action was captured
by the use of a brain monitoring system. The specific vignette used was the following
(Better condition in parenthesis and italics):
Mary is the single mother of two: Mark, 7, Sally, 4. Mary works most of the
day, and although she is known for being fairly patient and good-natured, lately
she has been under a lot of stress. Due to financial trouble, many of her co-
workers are being laid off, while the remaining ones are getting salary cuts. As a
result, she has been taking on many more responsibilities and working extra-
hours. Yesterday, when she came back from work late at night, she couldnt
drive into her garage because her neighbor had blocked her driveway with his
new BMW. Enraged, she stepped on the gas pedal and smashed her car into her
neighbors. Unfortunately, her neighbor was still inside the car (it was too dark
for anyone to see him), and both his legs were seriously broken in several
places. Now he is not only suing her for several thousand dollars, but hes also
pressing charges. Coincidentally, while all this happened, Mary was wearing a
BrainCap©: a new device that works like a heart monitor but instead of
recording the activity of your heart, it records the activity of your brain. In
fact, it records all electrical and chemical activity going on in your brain during
a particular period of time. Mary was wearing the BrainCap© because she was a
participant in a research study at the universityit was a good way of getting
some extra cash. The neuroscientist in charge of the study was then called to
read the brain activity which occurred during the event, and the result was rather
clear: right after seeing the parked BMW, and right when she was about to press
the gas pedal after seeing the BMW, Marys motor cortex (i.e. the area of the
brain controlling muscular movements) was active while her pre-frontal cortex
was inactive. This indicates, said the neuroscientist, that her frontal lobes
were unable to suppress the emotions she was feeling, and as a result they could
not inhibit the impulse to press the pedal. The pre-motor cortex what then left
uninhibited, and it sent the instruction to her foot which led to Marys pedal
pressing. Since there was no brain activity in her frontal lobe, which is required
to control emotions, at that moment Mary was completely unable to control her
rage and her desire to crash the car. In fact, any person with this kind of brain
activity would have done exactly the same thing. The set of brain events that led to
the pedal pressing was completely determined; Mary just couldnthavedone
otherwise. If Mary is found responsible for her actions, she may be sent to a
federal medical facility for the next 6 months. There she could receive medical
treatment, but she wont be able to see her children. Unfortunately [Fortunately],
during that time, they would be living with Social Services [Aunt Elizabeth], in
what might be a much worse (better) environment for them.
264 F. Brigard, W.J. Brady
3.2 Results and Discussion
The results of Study 2 showed that participants were more likely to say that Mary was
responsible for running her car into her neighborsintheBetter condition (M=5.75; SD =
1.26) than they were in the Worse condition (M=4.38; SD=1.76),t(23) =12.145, p<.001,
r=.41 (Fig. 1b). These results were consistent with those obtained in Study 1: participants
were more likely to attribute responsibility in the case of a third party possibly being
better off as a result of holding the agent responsible than they were if the third party
could end up worse off. However, unlike Study 1, participants assigned to the Worse
condition were no more likely to exculpate the agent, as their answers sat on the mid-
point. Nonetheless, these results still suggest that MRECD descriptions of deterministic
scenarios do not necessarily elicit compatibilist judgments of responsibility. After all,
when participants were confronted with the possibility of an innocent third party being
worse off, their responses did not differ from the mid-point, indicating uncertainty as to
whether or not responsibility should be ascribed.
In addition, these results lend further support to the suggestion that factors other
than the causal history preceding the agents action can influence peoples judgment
of responsibility. Whether or not future consequences of holding the agent responsi-
ble may make an innocent third party better or worse appear to differentially affect
peoples judgment of responsibility. Finally, these results also provide additional
support to the idea that moral considerations on possible future side effects that have
not yet occurred (as opposed to past side effects that have already occurred) can affect
peoples judgment of responsibility.
At this point it is reasonable to wonder whether the effect of possible future
consequences is a more pervasive feature of our judgments of responsibility in
general, regardless of whether or not the description of the causal history leading
up to the to-be-evaluated action is described in MRECD terms. Moreover, it is
reasonable to wonder whether this effect influences our judgments of responsibility
even in the absence of a description of the events preceding the action upon which the
agent is evaluated. To explore this question a third study was conducted.
4 Study 3
4.1 Procedure
Forty-eight undergraduates from UNC-Chapel Hill (Age M= 20.9, SD = 3.9) partici-
pated in this study. The methodology was identical to that in Studies 1 and 2, except
that this time all references to the deterministic character of the events prior to the
agents action were left out. The specific vignette used was the following (Better
condition in parenthesis and italics):
Mary is the single mother of two: Mark, 7, Sally, 4. Mary works most of the
day, and although she is known for being fairly patient and good-natured, lately
she has been under a lot of stress. Due to financial trouble, many of her co-
workers are being laid off, while the remaining ones are getting salary cuts. As a
result, she has been taking on many more responsibilities and working extra-
The Effect of What We Think 265
hours. Yesterday, when she came back from work late at night, she couldnt
drive into her garage because her neighbor had blocked her driveway with his
new BMW. It was cold, very dark, and she was exhausted. Enraged, she stepped
on the gas pedal and smashed her car into her neighbors. Unfortunately, her
neighbor was still inside the car (it was too dark for anyone to see him), and his
leg was broken. Now he is not only suing her for several thousand dollars, but
hes also pressing charges. If Mary is found responsible for her actions, she may
be sent to a correctional facility for the next 6 months, and she wont be able to
see her children. Unfortunately [Fortunately], during that time, they would be
living with their alcoholic father [Aunt Elizabeth], in what might be a much
worse [better] environment for them.
2
4.2 Results and Discussion
The results of Study 3 revealed that participants were more likely to say that Mary
was responsible for running her car into her neighbors in the Better condition (M=
6.17; SD=1.24) than they were in the Worse condition (M=4.46; SD = 2.21), t(22) =
9.900, p<.001, r=.43 (Fig. 1c). These results are consistent with those from Studies 1
and 2, as participants were more likely to attribute responsibility to Mary if a third
party could be better off than if a third party could be worse off in the future.
However, unlike study 1, and like study 2, participants were no more likely to
exculpate Mary when a third party could be worse off in the future, as their responses
did not differ from the mid-point.
Consistent with the results in Studies 1 and 2, the results of Study 3 suggest that
MRECD descriptions of deterministic scenarios do not necessarily elicit compatibilist
judgments of responsibility. Once again, when participants were confronted with the
possibility of an innocent third party being worse off, their responses indicated
uncertainty as to whether or not responsibility should be assigned. Moreover, these
results strengthen the suggestion that factors other than the causal history preceding
the agents action influence peoples judgment of responsibility. Finally, these results
lend further support to the claim that moral considerations on possible future side
effectsas opposed to past side effects that have already occurredcan affect
peoples judgment of responsibility.
5 General Discussion
The current studies manipulated considerations about possible future consequences
that may follow as a result of holding an agent responsible in order to investigate
three inter-related issues. First, we wanted to know whether MRECD descriptions of
2
This study was initially conducted using the same wording for the Worse condition as in the previous two
vignettes, so the last line read Unfortunately, during that time, they would be living with Social Services,
in what might be a much worse environment for them. This study, however, did not yield a significant
effect (p=.08) with an equivalent number of participants. Although the non-significant result in this first
study could have been a power issue, we decided to conduct the study anew changing the wording of the
Worse condition, so instead of Social Servicesit read, as shown above, alcoholic father. This variation
allows us to verify that the effect translates into other possible future negative consequences.
266 F. Brigard, W.J. Brady
the causal history preceding an action generate compatibilist judgments of responsi-
bility unconditionally. The results from the studies reported above suggest that the
alleged concrete system does not operate unconditionally, as even in the face of
MRECD descriptions, which previous studies have shown to elicit compatibilist
judgments of responsibility (e.g., Nichols and Knobe 2007; Nichols 2011), consid-
erations about possible future consequences affect peoples responsibility judgments.
Second, and more generally, the current studies investigated whether ordinary
people are affected by other considerations besides the causal history of the events
preceding the action evaluated. The three studies suggest that people may take into
account not only the causal history, but also what could happen in the future if the
agent is held responsible at a present time. The particular manipulation employed
here showed that if holding an agent responsible could bring about a possible future
undesirable consequence for an innocent third party, participants were less inclined to
attribute responsibility to that agent than if the consequence was desirable.
Finally, by specifically manipulating possible future side effects of holding an
agent responsible, these studies attempted to shed light on whether the side-effect
effect (Knobe 2003; Feltz 2007) only applies to side-effects that have occurred prior
to the moment in which the agent is judged. The results suggest that people factor in
possible future consequences of holding an agent responsible in a way that accords
with extant evidence on the side-effect effect. That is, people are more inclined to say
that the agent is responsible for the action when the future side effect is negatively
valued than when it is positively valued.
There was one important difference that emerged between the results in Study 1,
and Studies 2and 3. While in all three studies participants gave higher ratings of
responsibility when the possible future side effects were positively valued, only in
Study 1 were negatively valued possible future side-effects associated with partic-
ipantsexculpatory judgments. What could explain this asymmetric pattern of
results? One possibility is that, when participants read about the possible future
consequences on the innocent third party (i.e., Marys children), the perceived
valence of such a possible consequence determined whether or not it was to be
factored in their judgment of responsibility. Positively valued future side effects
appeared not to be factored in, and so they did not affect the output of the alleged
concrete system, which led to compatibilist judgments of responsibility. On the other
hand, negatively valued outcomes appear to be sometimes factored in, and thus are
capable of overriding the concrete systems response by allowing participants to
reassess their judgments of responsibility.
If this account is on the right track, then the results of Study 1 may be explained in
the following manner: positively valued possible future side effects did not affect the
performance of the concrete system, and thus judgments of responsibility in line with
compatibilism were elicited. However, negatively valued possible side effects did
override the operations of the system, allowing the subject to re-focus her attention on
other critical features of the vignettelike its MRECD featuresprior to making the
judgment. Consequently, in the Worse condition, participantsjudgments may have
been switched from compatibilism to incompatibilism either because the pathological
component of the description elicited an intense affective reaction (Greene 2007), or
because it lessened the impulse to blame (Knobe 2005) or punish (Nadelhoffer 2006)
the agent. In Studies 2 and 3, however, the perceived negative value of the possible
The Effect of What We Think 267
future consequences was only strong enough to override the normal operation of the
concrete system, but none of the MRECD features of the description was sufficient to
swing the judgment toward incompatibilism. As such, participantsresponses hov-
ered around the mid-point. At any rate, even if the suggested explanation is incom-
plete or inaccurate, we believe that the results reported here suggest that the behavior
of the alleged concrete system for judgments of responsibility is more complex than
originally thought. Whether or not similar manipulations may affect the abstract
system is an interesting question for future research.
However, it is important to acknowledge a number of interpretative limitations to
the results of the studies reported here. First, the way in which participants are asked
to register their responses in each vignette may be subject to more than one poten-
tially diverging reading. For instance, although participants were invited to read the
question as asking them to judge the degree to which the protagonist is morally
responsible, they may have read it instead as asking them to judge whether or not they
think it is appropriate to hold her responsible for the relevant action. We believe that
further research is needed to adjudicate between these two readings in order to clarify
the precise nature of the judgment upon which considerations about what may happen
in the future actually have an impact.
Another potential difficulty is that participants may have interpreted the question
as asking them to judge whether or not the protagonist deserved to be held respon-
sible, or whether or not the protagonist was blameworthy, or even whether or not the
protagonist should be punished.
3
Given the many purposes for which people normal-
ly employ the concept of moral responsibility, and the variety of situations in which it
gets to be deployed, the rather exploratory character of the studies reported here is
insufficient to adjudicate between all these potentially divergent readings of the
vignettesquestions. Clearly, further research is needed to shed light on these issues.
Nonetheless, we believe that the results reported here are interesting in their own
right, and that they offer evidence to the effect that, when it comes to studying the
folks concept of moral responsibility, participants considerations about possible
future consequences should not be overlooked.
References
Beebe, J., and W. Buckwalter. 2010. The epistemic side-effect effect. Mind & Language 25(4): 474498.
Björnsoon, G., and Persson, K. In press. A unified empirical account of responsibility judgments.
Philosophy and Phenomenological Research. doi:10.1111/j.1933-1592.2012.00603.x
De Brigard, F., E. Mandelbaum, and D. Ripley. 2009. Responsibility and the brain sciences. Ethical Theory
and Moral Practice 12(5): 511524.
Feltz, A. 2007. The knobe effect: A brief overview. Journal of Mind and Behavior 28: 265277.
Fischer, J.M. 1994. The metaphysics of free will. Oxford: Blackwell Publishers.
Greene, J. D. 2007. The secret joke of Kants soul. In Moral psychology, Vol. 3: The neuroscience of
morality: Emotion, disease, and development, ed. W. Sinnott-Armstrong. Cambridge, MA: MIT Press.
Greene, J.D., and J.D. Cohen. 2004. For the law, neuroscience changes nothing and everything.
Philosophical Transactions of the Royal Society B 359: 177517785. Special Issue on Law and
the Brain.
3
We thank an anonymous reviewer for pointing out these potential confounds.
268 F. Brigard, W.J. Brady
Kane, R. 1996. The significance of free will. New York: Oxford University Press.
Knobe, J. 2003. Intentional action and side effects in ordinary language. Analysis 63(3): 190194.
Knobe, J. 2005. Cognitive processes shaped by the impulse to blame. Brooklyn Law Review 71: 929937.
Knobe, J., W. Buckwalter, P. Robbins, H. Sarkissian, T. Sommers, and S. Nichols. 2011. Experimental
philosophy. Annual Review of Psychology 63(50): 7273.
Nadelhoffer, T. 2006. Bad acts, blameworthy agents, and intentional actions: Some problems for jury
impartiality. Philosophical Explorations 9(2): 203220.
Nahmias, E.A., S.G. Morris, T. Nadelhoffer, and J. Turner. 2005. Surveying freedom: Folk intuitions about
free will and moral responsibility. Philosophical Psychology 18(5): 561584.
Nahmias, E., S.G. Morris, T. Nadelhoffer, and J. Turner. 2006. Is incompatibilism intuitive? Philosophy and
Phenomenological Research 73(1): 2853.
Nahmias, E., D.J. Coates, and T. Kvaran. 2007. Free will, moral responsibility, and mechanism:
Experiments on folk intuitions. Midwest Studies in Philosophy V. XXXI: 214242.
Nichols, S. 2011. Experimental philosophy and the problem of free will. Science 331: 14011403.
Nichols, S., and J. Knobe. 2007. Moral responsibility and determinism: The cognitive science of folk
intuitions. Nous XLI(4): 663685.
Pereboom, D. 2001. Living without free will. Cambridge: Cambridge University Press.
Roskies, A., and S. Nichols. 2008. Bringing responsibility down to earth. Journal of Philosophy CV(7):
371388.
Sarkissian, H., A. Chatterjee, F. De Brigard, J. Knobe, S. Nichols, and S. Sirker. 2010. Is belief in free will a
cultural universal? Mind & Language 25(3): 346358.
Sinnott-Armstrong, W. 2008. Abstract + Concrete = Paradox. In Experimental philosophy, ed. J. Knobe and
S. Nichols, 209230. New York: Oxford University Press.
Woolfolk, R.L., J.M. Doris, and J.M. Darley. 2006. Identification, situational constraint, and social
cognition: Studies in the attribution of moral responsibility. Cognition 100: 283301.
The Effect of What We Think 269
Article
Full-text available
Over the past few decades, psychology and its cognate disciplines have undergone substantial scientific reform, ranging from advances in statistical methodology to significant changes in academic norms. One aspect of experimental design that has received comparatively little attention is incentivization, i.e., the way that participants are rewarded and incentivized monetarily for their participation in experiments and surveys. While incentive-compatible designs are the norm in disciplines like economics, the majority of studies in psychology and experimental philosophy are constructed such that individuals’ incentives to maximize their payoffs in many cases stand opposed to their incentives to state their true preferences honestly. This is in part because the subject matter is often self-report data about subjective topics, and the sample is drawn from online platforms like Prolific or MTurk where many participants are out to make a quick buck. One mechanism that allows for the introduction of an incentive-compatible design in such circumstances is the Bayesian Truth Serum (BTS; Prelec, 2004 ), which rewards participants based on how surprisingly common their answers are. Recently, Schoenegger (2021) applied this mechanism in the context of Likert-scale self-reports, finding that the introduction of this mechanism significantly altered response behavior. In this registered report, we further investigate this mechanism by (1) attempting to directly replicate the previous result and (2) analyzing if the Bayesian Truth Serum’s effect is distinct from the effects of its constituent parts (increase in expected earnings and addition of prediction tasks). We fail to find significant differences in response behavior between participants who were simply paid for completing the study and participants who were incentivized with the BTS. Per our pre-registration, we regard this as evidence in favor of a null effect of up to V = .1 and a failure to replicate but reserve judgment as to whether the BTS mechanism should be adopted in social science fields that rely heavily on Likert-scale items reporting subjective data, seeing that smaller effect sizes might still be of practical interest and results may differ for items different from the ones we studied. Further, we provide weak evidence that the prediction task itself influences response distributions and that this task’s effect is distinct from an increase in expected earnings, suggesting a complex interaction between the BTS’ constituent parts and its truth-telling instructions.
Article
Full-text available
A key challenge in experimental social science research is the incentivisation of subjects such that they take the tasks presented to them seriously and answer honestly. If subject responses can be evaluated against an objective baseline, a standard way of incentivising participants is by rewarding them monetarily as a function of their performance. However, the subject area of experimental philosophy is such that this mode of incentivisation is not applicable as participant responses cannot easily be scored along a true-false spectrum by the experimenters. We claim that experimental philosophers’ neglect of and claims of unimportance about incentivisation mechanisms in their surveys and experiments has plausibly led to poorer data quality and worse conclusions drawn overall, potentially threatening the research programme of experimental philosophy in the long run. As a solution to this, we propose the adoption of the Bayesian Truth Serum, an incentive-compatible mechanism used in economics and marketing, designed for eliciting honest responding in subjective data designs by rewarding participant answers that are surprisingly common. We argue that the Bayesian Truth Serum (i) adequately addresses the issue of incentive compatibility in subjective data research designs and (ii) that it should be applied to the vast majority of research in experimental philosophy. Further, we (iii) provide an empirical application of the method, demonstrating its qualified impact on the distribution of answers on a number of standard experimental philosophy items and outline guidance for researchers aiming to apply this mechanism in future research by specifying the additional costs and design steps involved.
Article
Full-text available
Skeptical worries about moral responsibility seem to be widely appreciated and deeply felt. To address these worries—if nothing else to show that they are mistaken—theories of moral responsibility need to relate to whatever concept of responsibility underlies the worries. Unfortunately, the nature of that concept has proved hard to pin down. Not only do philosophers have conflicting intuitions; numerous recent empirical studies have suggested that both prosaic responsibility judgments and incompatibilist intuitions among the folk are influenced by a number of surprising factors, sometimes prompting apparently contradictory judgments. In this paper, we show how an independently motivated hypothesis about responsibility judgments provides a unified explanation of the more important results from these studies. According to this ‘Explanation Hypothesis’, to take an agent to be morally responsible for an event is to take a relevant motivational structure of the agent to be part of a significant explanation of the event. We argue that because of how explanatory interests and perspectives affect what we take as significant explanations, this analysis accounts for the puzzling variety of empirical results. If this is correct, the Explanation Hypothesis also provides a new way of understanding debates about moral responsibility.
Article
Full-text available
Recent experimental research has revealed surprising patterns in people's intuitions about free will and moral responsibility. One limitation of this research, however, is that it has been conducted exclusively on people from Western cultures. The present paper extends previous research by presenting a cross-cultural study examining intuitions about free will in subjects from the United States, Hong Kong, India and Colombia. The results revealed a striking degree of cross-cultural convergence. In all four cultural groups, the majority of participants said that (a) our universe is indeterministic and (b) moral responsibility is not compatible with determinism.
Article
Full-text available
Philosophers working in the nascent field of 'experimental philosophy' have begun using methods borrowed from psychology to collect data about folk intuitions concerning debates ranging from action theory to ethics to epistemology. In this paper we present the results of our attempts to apply this approach to the free will debate, in which philosophers on opposing sides claim that their view best accounts for and accords with folk intuitions. After discussing the motivation for such research, we describe our methodology of surveying people's prephilosophical judgments about the freedom and responsibility of agents in deterministic scenarios. In two studies, we found that a majority of participants judged that such agents act of their own free will and are morally responsible for their actions. We then discuss the philosophical implications of our results as well as various difficulties inherent in such research.
Chapter
Experimental philosophy is a new movement that seeks to return the discipline of philosophy to a focus on questions about how people actually think and feel. Departing from a long-standing tradition, experimental philosophers go out and conduct systematic experiments to reach a better understanding of people’s ordinary intuitions about philosophically significant questions. Although the movement is only a few years old, it has already sparked an explosion of new research, challenging a number of cherished assumptions in both philosophy and cognitive science. The present volume provides an introduction to the major themes of work in experimental philosophy, bringing together some of the most influential articles in the field along with a collection of new papers that explore the theoretical significance of this new research.
Article
Joshua Knobe (2003a) has discovered that the perceived goodness or badness of side effects of actions influences people's ascriptions of intentionality to those side effects. I present the paradigmatic cases that elicit what has been called the Knobe effect and offer some explanations of the effect. 1 put these explanations into two broad groups. One explains the Knobe effect by referring to our concept of intentional action. The other explains the Knobe effect without referring to our concept of intentional action. I discuss some problems with these explanations and conclude with some possible avenues for future research.
Article
Knobe (2003a, 2003b, 2004b) and others have demonstrated the surprising fact that the valence of a side-effect action can affect intuitions about whether that action was performed intentionally. Here we report the results of an experiment that extends these findings by testing for an analogous effect regarding knowledge attributions. Our results suggest that subjects are less likely to find that an agent knows an action will bring about a side-effect when the effect is good than when it is bad. It is further argued that these findings, while preliminary, have important implications for recent debates within epistemology about the relationship between knowledge and action.
Article
Most people assume that, even though some degenerative or criminal behavior may be caused by influences beyond our control, ordinary human actions are not similarly generated, but rather are freely chosen, and we can be praiseworthy or blameworthy for them. A less popular and more radical claim is that factors beyond our control produce all of the actions we perform. It is this hard determinist stance that Derk Pereboom articulates in Living Without Free Will. Pereboom argues that our best scientific theories have the consequence that factors beyond our control produce all of the actions we perform, and that because of this, we are not morally responsible for any of them. He seeks to defend the view that morality, meaning and value remain intact even if we are not morally responsible, and furthermore, that adopting this perspective would provide significant benefit for our lives.