Ironic Effects of Mental Control in Problem Solving: Evidence for the Implementation of Ineffective Strategies

Abstract

The ironic effect of intending to solve problems was examined in this study. Previous research has demonstrated ironic effects of mental control for numerous behavioral and cognitive processes. In this study, subjects were either asked to solve problems, or they were asked to solve the same problems as quickly and efficiently as possible. Based on previous demonstrations of ironic effects of mental control, it was expected that those exercising the greatest mental control would have the poorest performance. Results indicated that those subjects trying to solve problems quickly and efficiently actually solved fewer problems and committed more errors than those who were not intended to work quickly and efficiently. Furthermore, there is evidence that this ironic effect of mental control in problem solving was associated with the use of different strategies. The use of ineffective strategies is suggested as one explanation for the ironic effects of mental control.

Full text

AMERICAN JOURNAL OF
PSYCHOLOGICAL RESEARCH
Volume 2, Number 1
Submitted: September 12, 2006
Accepted: November 1, 2006
Publication Date: November 15, 2006
Ironic Effects of Mental Control in Problem Solving: Evidence for the
Implementation of Ineffective Strategies
Christian L. Hart & Joe A. Randell
East Central University
Christian L. Hart, Department of Psychology, East Central University, Ada, Oklahoma 74820,
USA. E-mail: chart@mailclerk.ecok.edu
ABSTRACT
The ironic effect of intending to solve problems was examined in this study. Previous research
has demonstrated ironic effects of mental control for numerous behavioral and cognitive
processes. In this study, subjects were either asked to solve problems, or they were asked to solve
the same problems as quickly and efficiently as possible. Based on previous demonstrations of
ironic effects of mental control, it was expected that those exercising the greatest mental control
would have the poorest performance. Results indicated that those subjects trying to solve
problems quickly and efficiently actually solved fewer problems and committed more errors than
those who were not intended to work quickly and efficiently. Furthermore, there is evidence that
this ironic effect of mental control in problem solving was associated with the use of different
strategies. The use of ineffective strategies is suggested as one explanation for the ironic effects
of mental control.
INTRODUCTION
The theory of ironic processes of mental control suggests that in some situations, the
process of intending to take some action can ironically lead to counter-intentional behaviors
(Wegner, 1994). That is, by intending to do something, we sometimes find ourselves doing just
the opposite. These ironic effects of mental control have recently been demonstrated in a number
of domains, including thought suppression (Wegner & Zanakos, 1994), relaxation (Wegner,
Broome, & Blumberg, 1997), sleep onset (Ansfield, Wegner, & Bowser, 1996), and motor
40

behaviors (Wegner, Ansfield, & Pilloff, 1998). In the present study, ironic effects of intention
were investigated for subjects engaged in problem solving tasks.
According to Wegner & Pennebaker (1993, p. 1), “Mental control occurs when people
suppress a thought, concentrate on a sensation, inhibit an emotion, maintain a mood, stir up a
desire, squelch a craving, or otherwise exert influence on their own mental states.” Quite a few
studies have established that mental control can backfire, leading to cognitive or behavioral
outcomes opposite of those that were intended. For example, people trying to fall asleep quickly
tend to stay awake longer (Ansfield, Wegner, & Bowser, 1996). When people try to relax, they
tend to produce elevated physiological indications of anxiety (Wegner, Broome, & Blumberg,
1997). Wegner and Zanakos (1994) found that when people try to suppress particular thoughts,
they tend to find those very thoughts more accessible or intrusive. Recently, research in this lab
revealed that when people tried hard to remember information, they tended to remember less
than people who were not trying hard to remember (Hart & Randell, 2006). In the
aforementioned studies, ironic effects of mental control typically arose when subjects were
placed under conditions of elevated cognitive workload. While ironic effects do not occur every
time mental control is exercised, they are certainly not isolated to a narrowly defined set of
circumstances.
The concept of ironic effects of mental control is not new to psychology. Over a century
ago, Freud (1901/1938) noted that an attempt to carry out one action can result in the production
of an opposite or counter-intentional action. Wegner (1994) recently termed these failures of
mental control that are in opposite directions of one’s intended actions ironic effects of mental
control. Wegner suggested that ironic effects of mental control tend to occur when high levels of
workload cause mental control strategies to fail. Wegner’s ironic process theory proposes that as
mental control strategies fail, cognitive monitoring systems, operating outside of conscious
awareness, increase the probability that effects opposite of those intended will be generated
(Wegner, 1994). For example, if a person is trying to suppress a thought, there will be a
conscious strategy employed to suppress the thought, and there will be a subconscious
monitoring system that searches for evidence that the strategy is failing. That is, the monitor
searches for evidence of the unwanted thought. According to Wegner, the conscious mental
control systems sometimes fail when people are experiencing high cognitive workload. As these
control systems fail, the subconscious monitoring system leads to the hyperaccessibility of the
unwanted thought. Thus, the overall system failure leads to the ironic hyperaccessibility of the
very thoughts one is trying to forget.
In this lab, we recently developed an alternative theory to account for ironic effects of
mental control that we call ineffective strategy implementation theory (ISI). ISI suggests that
when individuals exercise high mental control, they adopt new strategies in an attempt to alter
their performance. These new strategies may be unproven, less effective, and more likely to fail
as cognitive workload increases. As a consequence, high mental control often leads to the
implementation of ineffective strategies, which leads to poor performance. While ISI has not
been thoroughly examined, it does appear to be capable of accounting for the results reported in
recent studies of ironic effects of mental control.
41

This study investigated whether ironic effects of mental control occur when subjects are
consciously working hard to solve problems. Based on previous studies of ironic effects, it was
expected that individuals who tried the hardest to solve problems would experience the most
pronounced ironic failures. The hypothesis of this experiment was that a high level of intent to
solve problems (high mental control) would be less effective than a lower level of intention to
solve problems. Specifically, it was expected that an attempt to solve problems as quickly and
efficiently as possible would result in poorer performance than if subjects were not attempting
such a quick and efficient performance.
Another aim of this study was to determine if, in fact, subjects exercising high mental
load appeared to utilize different strategies than those employing lower levels of mental control.
According to ISI theory, there should be evidence that those subjects under high mental control
use different strategies. In addition to performing worse, evidenced by solving fewer problems
and committing more errors, it was believed that ironic failures would be associated with
evidence of subjects using different strategies in their attempt to solve problems.
METHOD
Participants
Fifty-four male and female undergraduate students from introductory psychology courses
served as participants in this experiment. Participation was voluntary. Informed consent was
obtained from all participants in this study.
Design and Procedure
All participants were told that they would be attempting to solve word puzzle problems.
Participants were randomly assigned to one of two groups. The two groups represented the two
levels of the independent variable, mental control. The two groups differed in the problem
solving instructions they were given. One group was told to simply solve the problems (low
mental control), while the other group was instructed to solve as many problems as quickly and
efficiently as they could (high mental control).
All participants were engaged in a word scramble task that required them to solve
problems by rearranging four letter combinations in order to create words. For example, if
participants were presented with the word scramble, OWMR, they would need to unscramble the
letters and spell the word, WORM, on a sheet of paper. Similarly, if they were presented with the
word scramble, CKDU, they would need to unscramble the letters and spell the word, DUCK, on
the sheet of paper. All participants were given two examples of word scramble problems and
their correct solutions prior to testing in order to ensure that all subjects understood the problem
solving task. Participants were given a sheet of paper with forty-six of these word scramble
problems. All participants were given three minutes to work on the word scramble problems,
however, they were not told that a time limit was in place.
The three dependent measures for which data were collected were the number of
correctly solved word scramble puzzles, the number of incorrectly solved word scramble
puzzles, and the number of word scramble puzzles that were skipped. Correct solutions were
42

those that spelled out words found in Webster’s New World dictionary (1984). Incorrect
solutions were those that did not appear in the dictionaries. A skipped problem was defined as
any word scramble problem for which a subject response was not provided, followed by a
subsequent word scramble problem for which a subject response was provided. Correct and
incorrect solutions were used to determine whether or not there had been ironic effects of mental
control. Skipped problems allowed us to explore whether or not the participants in the two
groups appeared to employ different strategies as suggested by ISI theory. This variable was
selected as a result of observations made during a pilot study. It was observed that participants in
the high mental control condition spent less time working on a problem before giving up and
skipping ahead to another problem.
RESULTS
The Hotelling’s Trace multivariate test indicated that there was a significant difference
between the two sample groups, F(3, 50) = 3.923, p < .05. Univariate tests (independent t-tests)
were used to examine group differences for each dependent variable.
The results indicated that there was a significant difference between the two groups in the
number of word scramble puzzles solved, t(52) = 2.537, p < .01 (one-tailed), with subjects in the
high mental control condition solving significantly fewer problems (M = 18.62, SEM = 1.96)
than subjects who were in the low mental control condition (M = 25.15, SEM = 1.66). An
analysis of data for the number of incorrect solutions revealed a significant difference between
groups, t(52) = 1.774, p < .05 (one-tailed), with those in the high mental control condition
generating more incorrect responses (M = 2.19, SEM = .68) than participants in the low mental
control condition (M = .89, SEM = .27). An incorrect solution was defined as a solution to a
word scramble puzzle that did not appear in English language dictionaries. These results are
consistent with the prediction that ironic effects of mental control would occur when people tried
to solve problems quickly and efficiently.
A third univariate analysis explored whether or not the subjects in the two groups
appeared to employ different strategies as suggested by ISI theory. The specific hypothesis was
that participants in the high mental control condition would skip more problems than those in the
low mental control condition. An analysis of the number of problems skipped revealed a
significant difference between groups, t(52) = 1.706, p < .05 (one-tailed), with participants in
the high mental control condition skipping more problems (M = 10.07, SEM = 1.81) than
participants in the low mental control condition (M = 6.22, SEM = 1.36). These results are
consistent with the prediction that conditions associated with the production of ironic effects
would also be associated with the use of different strategies.
DISCUSSION
The results of this study suggest that there are ironic effects of mental control for problem
solving tasks. This finding adds to a growing body of evidence indicating that people exercising
high levels of mental control experience ironic failures when compared to their counterparts who
adopt less performance-driven approaches. Additionally, the apparent use of different strategies
by individuals in this study is consistent with Ineffective Strategy Implementation theory. That
43

is, ironic failures associated with high levels of mental control may be due to the selection and
implementation of strategies that are less than optimal for the task at hand.
Wegner (1994) suggested that ironic effects of mental control occur when conscious
control mechanisms (i.e., the strategies) become overwhelmed by high cognitive workload, and
an unconscious monitoring system continues to operate as the control system falters. According
to Wegner, this monitoring system operating outside of conscious awareness begins to introduce
thoughts of unsuccessful performance. These unconscious thoughts of the behaviors one is
attempting to avoid begin to activate those very behaviors one had been attempting to evade. In
the present case, Wegner’s hypothetical unconscious monitor would be a system that constantly
looked for examples of problems being solved slowly and inefficiently. The theory suggests that
once the controlled problem solving processes began to fail, this unconscious monitor caused
problem solving errors to occur. In the case of the present study, Wegner’s theory can account
for our observations; however, his theory does not account for the apparent differences in
strategy between the two groups.
The results of this study, including the variations in strategy, do fit well with ISI theory.
It seems that in many cases of ironic failures, people may simply be selecting and implementing
ineffective strategies. For example, if people work hard to fall asleep quickly, they may adopt
strategies such as counting sheep. As it turns out, these strategies might just increase mental
activity, thereby ensuring that the person will stay awake longer than a person who has not
implemented any conscious strategies whatsoever. Our ineffective strategy implementation
theory suggests that when individuals are given some instructional cue to improve performance
beyond levels with which they are accustomed, those individuals feel compelled to do something
in addition to or different from what they typically do. They adopt a new strategy in order to
affect such a change in performance. When we make an effort to exercise mental control over
our problem solving capabilities, we inevitably call upon some strategy. That is, we try to do
something in order to work faster. Because these novel strategies are untested, there is a chance
that they are not wholly effectual. It may be that these ineffective strategies are selected without
any explicit awareness on the part of the person. Cary and Reder (2002) noted that subjects select
strategies and shift strategies without conscious awareness of the strategies or the factors that
influence one’s strategies.
Changes in cognitive workload may also lead to ironic failures of mental control. The
application of new strategies necessitates shifting away from automatic processes toward more
controlled processes. Controlled processes, by their nature, generate a greater cognitive workload
than automatic processes (Jacoby, Yonelinas, & Jennings, 1997; Schneider & Shiffrin, 1977). It
may be that these increases in cognitive workload contribute to the observed ironic effects. As
people adopted controlled strategies in an attempt to increase problem performance, the
additional cognitive workload burden might have depleted limited resources, causing a
detectable drop in the pace of problem solving. Numerous studies have demonstrated that
increases in cognitive workload can lead to such a decline in overall cognitive performance
(Grasha, & Schell, 2001; Wickens, 1989). Further research will need to be conducted before any
conclusion can be reached about the accuracy and value of ISI theory. However, this theory does
appear to offer a plausible mechanism for ironic failures of mental control.
44

The outcomes of this study suggest that at times our attempts to expedite problem solving
behaviors may result in poorer performance than one would expect otherwise. It might be that
these ironic failures in problem solving are due to the selection of suboptimal strategies. Thus, by
attempting to do a better job, we may be ironically performing actions that directly lead to
inadequate performance or failure.
REFERENCES
Ansfield, M. E., Wegner, D. M., & Bowser, R. (1996). Ironic effects of sleep urgency. Behaviour
Research and Therapy, 34, 523-531.
Cary, M., & Reder, L. M. (2002). Metacognition in strategy selection. In P. Chambres, & M.
Izaute (Eds.), Metacognition: Process, function and use (pp. 63-77). Dordrecht,
Netherlands: Kluwer Academic Publishers.
Freud, S. (1938). Psychopathology of everyday life. In A. A. Brill (Ed. & Trans.), The basic
writings of Sigmund Freud (pp. 33-178). New York: Random House. (Original work
published 1901)
Grasha, A. F., & Schell, K. (2001). Psychosocial factors, workload, and human error in a
simulated pharmacy dispensing task. Perceptual & Motor Skills, 92, 53-71.
Hart, C. L., & Randell, J. A. (2006). Ironic effects of intending to remember.
Manuscript submitted for publication.
Jacoby, L. L., Yonelinas, A. P., & Jennings, J. M. (1997). The relation between conscious and
unconscious (automatic) influences: A declaration of independence. In Cohen, J. D., &
Schooler J. W. (Eds.), Scientific approaches to consciousness (pp. 13-48). Mahwah, NJ:
Erlbaum.
Schneider, W., & Shiffrin, R. M. (1977). Controlled and automatic human information
processing. I. Detection, search and attention. Psychological Review, 84(1), 1-66.
Webster’s new world dictionary. (1984). New York, NY: Simon & Schuster.
Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101, 34-52.
Wegner, D. M., Ansfield, M., & Pilloff, D. (1998). The putt and the pendulum: Ironic effects of
mental control in action. Psychological Science, 9, 196-199.
Wegner, D. M., Broome, A., & Blumberg, S. J. (1997). Ironic effects of trying to relax under
stress. Behaviour Research and Therapy, 35, 11-21.
Wegner, D. M., & Pennebaker, J. W. (1993). Changing our minds: An introduction to mental
control. In Wegner, D. M., & Pennebaker, J. W. (Eds.), Handbook of Mental Control
(pp. 1-12). Upper Saddle River, NJ: Prentice Hall.
Wegner, D. M., & Zanakos, S. (1994). Chronic thought suppression. Journal of Personality, 62,
615-640.
Wickens, C. D. (1989). Resource management and time sharing. In Elkind, J. I., Card, S. K.
(Eds.), Human Performance Models for Computer-Aided Engineering (pp. 180-202).
Washington, D.C.: National Academy Press.
45