Printer Friendly

Rule-Based insensitivity and delusion maintenance in schizophrenia.

Delusions constitute one of the core symptoms of schizophrenia and are highly frequent and varied. While the processes responsible for their development continue to be debated (Bentall, 1994; Garety & Freeman, 1999; Kinderman & Bentall, 1997, 2000; Maher, 1974, 1999), another central question is why delusions persist despite evident proof of their inaccuracy and of the negative consequences they often engender. Several cognitive mechanisms have been proposed to explain the maintenance of delusions in schizophrenia, including bias against disconfirmatory evidence, bias in favor of confirmatory evidence, and safety-seeking behaviors. These hypotheses point to cognitive processes but do not take into account the verbal nature of delusions. We will first review these three cognitive accounts. We will then focus on the verbal nature of delusions from a behavioral perspective, explain how it may be implicated in their maintenance, and present a pilot study to evaluate the influence of rule-following in delusion maintenance.

Cognitive Accounts to Explain Maintenance of Delusions

Bias Against Disconfirmatory Evidence

Several studies (Moritz & Woodward, 2006; Woodward et al. 2006a, b; Woodward et al. 2004; 2006a, b) have shown that chronic schizophrenia patients tend to ignore evidence inconsistent with their interpretations of the world. For example, when presented with interpretations of pictures that initially appeared to be true but eventually proved to be false after more information had been provided, participants with schizophrenia exhibited higher levels of bias against changing their interpretation as compared with control participants (Woodward et al., 2004; 2006a, b). This tendency of a bias against disconfirmatory evidence (BADE) was particularly pronounced for currently delusional patients and different from psychiatric controls. Moreover, BADE seems to occur independently of emotional valence since it has been observed in schizophrenia for affectively neutral material as well as for delusional ideas (Moritz & Woodward, 2006; Woodward et al. 2006a, b).

Bias in Favor of Confirmatory Evidence

Some authors (Freeman et al. 2002) have proposed that maintenance of delusions is based on a combination of BADE together with a process of obtaining confirmatory evidence for the belief. According to Freeman et al. (2002), in addition to normal confirmation bias involving seeking evidence consistent with one's beliefs, patients with schizophrenia also present attentional and memory biases in favor of confirmatory evidence, particularly with respect to threatening beliefs in the context of persecutory delusions.

A number of biases might contribute to confirmatory evidence in schizophrenia. First, a "jumping to conclusions bias" (Hemsley & Garety, 1986; Huq et al. 1988; Garety et al. 1991; Dudley et al. 1997) prevents patients from gathering more information about the situation, thus contributing to a memory bias for the already gathered confirmatory evidence. Second, studies on attributional bias have shown that patients tend to overattribute causal influence for positive and negative events to themselves and to external causes, respectively (Kaney & Bentall, 1989; Lyon et al. 1994; Sharp et al. 1997). These attributions could have the function of preserving self-esteem and avoiding the necessity to deal with a possible difference between the way they think they are and the way they would like to be (Kinderman & Bentall, 1996). This bias results in the externalization of negative events that confirm beliefs, notably persecutory ones. Third, people with schizophrenia often present a theory of mind dysfunction (Corcoran et al. 1997; Corcoran et al. 1995; Frith & Corcoran, 1996; Doody et al. 1998; Sarfati, 2000) leading to errors in reading the intentions of others in the direction of a confirmation of their initial interpretation.

According to this approach, the combination of these three biases (jumping to conclusions, attributional style, and theory of mind dysfunction) explains a bias in favor of confirmatory evidence, particularly for delusions of reference and persecution.

Safety Behaviors

Persecutory delusions often come with fear and actions intended to reduce perceived threats. Confronted with threat and anxiety, individuals with schizophrenia engage in avoidance, escape, or compliance responding (Startup et al. 2007; Freeman et al., 2007). This way of responding may be effective from the patients' point of view if no harmful consequence appears, and hence they may continue to indulge in it. However, by continuing to engage in such behaviors, they are also less likely to come into contact with disconfirmatory evidence for their delusional beliefs (Freeman et al., 2002; Freeman et al. 2001). Safety behaviors are thus reinforced, leading to maintenance and even support of delusion.

The same line of argument can be proposed for non-persecutory delusions, with these delusions also considered as safety behaviors. Non-persecutory delusions represent searches for meaning when confronted with perceptual anomalies and anomalous experiences (Maher, 1988). In these cases, any destabilization of the delusion might lead once again to distress and fear associated with anomalous perceptual experiences. Hence, these types of delusions might also be maintained because they provide an escape from fear.

Delusion Maintenance as an Effect of Rule-Based Insensitivity

The three accounts of delusion maintenance proposed so far assume a preservation of sensitivity to the environment. Biases against disconfirmatory evidence or in favor of confirmatory evidence involve checking the environment for evidence to confirm the belief or not. Similarly, safety-seeking behaviors imply a survey of the environment to categorize information as potentially unsettling or harmful. But the very fact that delusions are inflexible and resistant to new facts or arguments from others suggests insensitivity to the environment.

To date, the influence of verbal properties of delusions, notably the fact that language creates distance from some parts of the environment, has not been addressed. Delusions constitute verbal descriptions of the world as it is perceived and understood, sometimes in a very symbolic form. As such, delusions operate as verbal rules governing a range of behaviors consistent with them (e.g., "I mustn't look into people's eyes or they will read my thoughts"). Behavior analytic research may provide one possible route into investigating the verbal properties of delusions and their implications for insensitivity to some environmental change. Indeed, one possible cause of such insensitivity is rule-governed behavior, defined as a behavior, either verbal or nonverbal, under the control of verbal antecedents (Catania, 1991).

Extensive empirical research on rule-governed behavior shows that, when given an instruction (i.e., a rule) on how often to press a button to get as many reinforcers as possible, participants can become insensitive to subsequent changes in experimental consequences (Baron & Galizio, 1983, Catania et al. 1989, 1990; Hayes et al. 1986; Rosenfarb et al. 1992; Shimoff & Catania, 1998; Vaughan, 1989). For example, in Wulfert et al. (1994), participants were given accurate instructions on the correct pace at which to press two buttons connected with different schedules of reinforcement (i.e., a "multiple schedule") to earn as many points as possible. These participants were insensitive to subsequent changes in contingencies in comparison with participants who received no instructions. The latter adapted their responding, whereas the former failed to do so. In addition, studies on self-generated instructions have shown that "shaped" self-generated rules facilitate acquisition of behaviors in accordance to contingencies, but may be more likely to lead to insensitivity than instructed rules (e.g., Rosenfarb et al., 1992).

We hypothesize that exaggerated insensitivity to some environmental changes in case of rule-following constitutes what drives patients to delusional thinking. Once a rule about the functioning of the world, as well as the way to behave accordingly, has been emitted, one can usually adjust the rule and their behaviors according to the consequences of their behaviors. With inflated insensitivity to consequences of behaviors in case of rule-following, one may persist in inappropriate behaviors and keep any maladjusted rule. Hence, we hypothesize that patients who already presented with delusional ideas in the past may be more insensitive in case of rule-following. To test this hypothesis, we ran a pilot study to compare rule-based insensitivity to environmental changes in patients who formerly presented with delusional thinking and in control participants. We predicted that patients who had formerly presented with delusional thinking would demonstrate more insensitivity to changes of programmed contingencies of reinforcement than controls when following a rule, thus reflecting maintenance of delusions in this population. En addition, consistent with the fact that delusions are generally self-generated and particularly resistant to external counter arguments, we predicted that patients who had formerly presented with delusional thinking would be more insensitive to changes of programmed contingencies of reinforcement when following their own rule than when following a rule provided by another person.



The experimental group included 17 inpatients (13 males and 4 females) diagnosed with schizophrenia according to

IV criteria and entering the hospital for difficulties directly or indirectly related to delusions. Each inpatient presented delusions at their admission to the hospital. Patients participated in the experiment the day before they left the hospital, at a moment when their referent psychiatrist considered that delusions were absent, or at least sufficiently reduced to avoid social or occupational dysfunction. The mean age of this group was 35.5 (SD 13.7) years. Mean duration of illness was 8.7 (SD 6.6) years. All patients presented delusions during the course of the illness. Each patient was under antipsychotic medication when included in the experiment, with an average chlorpromazine equivalent dose of 390 mg. Patients gave written informed consent to participate in the study approved by the local ethics committee. (1)

The control group comprised 30 students presenting with no psychiatric diagnosis at the time of the experiment. This group included 22 females and 8 males with a mean age of 24.9 (SD 3.5) years.


The experiment was run on a P75-Mhz computer (32 MB RAM, HD 540 Mo) with a Windows 95(TM) operating system. The experimental software, created using the Visual Basic[TM] programming language, featured two on-screen buttons, both of which could be pressed (though only one at a time) using the space bar. A different reinforcement schedule was associated with each button. During the pretraining phase, only the left button was activated. In the experimental phase, the two buttons were activated during alternative 40-s periods, thus constituting a 40-s multiple schedule. An on-screen meter showed the number of points won by the participant (which varied according to schedule performance) during a particular experimental session.


The experiment took place in the first author's office in the hospital. The experimenter was present throughout the experiment in case of material difficulty.

Pretraining The instructions read to the participants before pretraining were as follows:
  "In this game, you can earn points by pressing this button
  with the space bar Your aim is to earn as many points as

In the pretraining phase, participants were exposed to tour different successive reinforcement schedules (fixed ratio 8, variable ratio 8, fixed interval 8, variable interval 8, each 4 min long. Participants were informed that they could earn points by pressing the onscreen button through the space bar and that their aim was to earn as many points as possible. The order of presentation of these four schedules was counterbalanced across participants. This baseline phase enabled comparison between patients and controls with respect to adjustment to schedules.

These types of schedules have been extensively studied. The patterns of responses typically obtained are an overall low rate of responding for interval schedules and rapid rate of responding for ratio schedules (Ferster 8c, Skinner, 1957).

Multiple Schedule After the completion of the pretraining phase, participants were exposed to two multiple schedule sessions of 4 min each. During these sessions, the two buttons worked during alternative periods of 40 s, for a total of three presentations of FR and three presentation of FI for each session, with the working and non-working buttons signaled by a green and a red color, respectively.

First, participants were exposed to a "regular" multiple schedule that involved an FR8 on the left button and an F18 on the right one.2 The initial instructions were the same as those presented during the pretraining phase. In addition, participants were told that they could operate the buttons at alternative times and that the activated button was the one signaled by the green color. After 4 min of the regular schedule, participants were exposed to 4 min of a "reversed" multiple schedule that was the exact opposite of the regular multiple schedule. This involved an FR8 on the right button and an FI8 on the left one. Participants were not informed of this change, however.

Groups Before exposure to the multiple schedule, participants in both the control and the patients groups were randomly assigned to one of three conditions, as follows: "With Instructions" (WI), "No Instructions" (NI), and "Self Instructions" (SI). This yielded a total of six groups: Controls WI (N=10), Controls NI (N= 10), Controls SI (N=10); Patients WI (N=6), Patients NI (N=5), and Patients SI (N=6).

Before the regular multiple schedule, the WI group was instructed as follows:
  "To earn as many points as possible on the 10 button, you
  have to press at a fast pace. To earn as many points as
  possible on the right button, you have to press at a slow
  and steady pace."

The WI group did not receive any further instructions throughout the experiment. Notably, the instructions were not repeated before the reversed multiple schedules.

No instructions were given to the NI group.

No instructions were given to the SI group before the regular multiple schedule, but after it finished (before the reversed phase), they were asked to complete the following sentences:
  "To earn as many points as possible on the left button, you

  "To earn as many points as possible on the right button, you

The working hypothesis for this choice of procedure is that asking participants to state their own rules on the functioning of each button would ensure that these rules are explicit to them and exhaustive regarding the schedules presented.

The whole experimental procedure is summarized in Table 1.
Table 1 Synthesis of the whole procedure

Group        Before        Regular       Before        Reversed
             presentation  multiple      presentation  multiple
             of regular    schedule      of reversed   schedule
             multiple                    multiple
             schedule                    schedule

With         Accurate      Alternation   [empty set]   Alternation
instruction  instructions  every 40 s                  every 40 s
             on how to     of Fixed                    of Fixed
             earn as many  Ratio 8                     Interval 8 s
             points as     (left                       (left
             possible on   button) -                   button)
             each          Fixed                       Fixed Ratio
             schedule      Interval 8 s                8 (right
                           (right                      button)

No           [empty set]                 [empty set]

Self         [empty set]                 Sentences to
instruction                              complete on
                                         how to earn
                                         as many
                                         points as
                                         possible on

For ethical reasons and in order for the participants to fully understand the experiment they participated in, they were told about, after they completed the experiment, the non-signaled changes that operated and the best way to earn as many points as possible on each button in regular and reversed parts of the experiment.


Pretraining Phase

The dependent variable was the total number of presses on the space bar. Table 2 shows the means and standard deviations for numbers of presses for both groups (i.e., controls and patients) and for each of the four schedules (FI, VI, FR, VR) in the pretraining phase. Ratio schedules usually produce higher rates of responding than interval schedules (Fetster & Skinner, 1957). Accordingly, Friedman ANOVAs showed that number of presses were significantly different across schedules, in favor of ratio schedules, for patients ([X.sup.2]=9.37, df=3; p<0.05) and for control ([X.sup.2]=38.65, df=3; p<0.05). This finding indicates the patients' sensitivity to the consequences of pressing across the different schedules.
Table 2 Mean and standard deviation for number of presses
per group and per schedule in pretraining phase

Schedule     Controls      Patients

            Mean   SD     Mean   SD
FI          508    355    522    286
VI          530    366    531    239
FR          905    371    703    292
VR          797    372    700    319

FT= fixed interval, VI--variable interval, FR= fixed ratio,
VR=variable ratio

Multiple Schedule

The total numbers of presses of each participant across schedules and groups are presented in Table 3. For each group (With Instruction, Self-Instruction, No Instruction), participants' sensitivity to schedule differences was evaluated by comparing the number of presses on the left and right buttons using Wilcoxon tests for two paired samples. A good adjustment was estimated if the difference between the numbers of presses was significant, with more presses on the FR schedule (left button in regular multiple schedule, right button in reversed multiple schedule). If no such difference was seen, then participants were judged as not having adjusted their responding to the consequences. A difference between FR and FI in the regular multiple schedule but not in the reversed multiple schedule would indicate that participants' behavior failed to adjust after the schedules were reversed without the participants being informed.
Table 3 Total numbers of presses of each participant across
schedules and groups

Participants  Number of  presses                       Group

              FR         FI        FR        FI
              regular    regular   reversed  reversed
              multiple   multiple  multiple  multiple
              schedule   schedule  schedule  schedule

Control 1       563        242       243       466      WI

Control 2       368        257       424       365      WI

Control 3       462         23       550        27      WI

Control 4       554         88       231       376      WI

Control 5       388        185       391       202      WI

Control 6       472        134       464       142      WI

Control 7       368         55       280        91      WI

Control 8       520         63       569        27      WI

Control 9       424        149       386       159      WI

Control 10      512         85       528       138      WI

Control 11      368        177       392       298      SI

Control 12      432         92       424       173      SI

Control 13      472         88       480       107      SI

Control 14      440        301       383       312      SI

Control 15      512        245       528       199      SI

Control 16      354        151       355        58      SI

Control 17      327        257       328        60      SI

Control 18      336        309       382       215      SI

Control 19      424         90       416        73      SI

Control 20      517        397       478       388      SI

Control 21      144         91        92        60      NI

Control 22      280         98       224       118      NI

Control 23      408        478       208       2I2      NI

Control 24      592         41       655        32      NI

Control 25      447        348       451       331      NI

Control 26      331        203       332        48      NI

Control 27      225         58       223        37      NI

Control 28      224        228       272       233      NI

Control 29      569         44       579        33      NI

Control 30      400        259       457        51      NI

Patient 1       264         32        22       304      WI

Patient 2       384        186       480        82      WI

Patienl 3       406        261       458       278      WI

Patient 4       216        109       128       159      WI

Patient 5       238         93       107        92      WI

Patienl 6       328        178       336       283      WI

Putient 7       384        271       392       159      SI

Patient 8        88         77        16        56      SI

Patient 9       256        195       287       212      SI

Patient 10      102         91       129        95      SI

Patient 11      400        328       424       306      SI

Patient 12      336        263        49       179      SI

Patient 13      304        306       384       272      NI

Patient 14      256        184       368       305      NI

Patient 15      453        323       465       292      NI

Patient 16      305        231       361       245      NI

Patient 17      336        359       464       420      NI

FR=fixed ratio, FI=fixed interval, WI=with instruction,
SI= self-instruction, NI=no instruction

For the first presentation of each schedule, Wilcoxon tests for two paired samples showed that the only significant differences appeared for the controls and the patients to whom an instruction was given (controls WI and patients WI; p<0.05). Discriminating between programs during these first contacts with each schedule was quite difficult for other participants. For this reason, and in order to study stabilized patterns of behaviors, the results have been analyzed without the first presentation of each schedule.

Table 4 shows the p values (Wilcoxon tests for two paired samples) of the differences between the mean number of presses on FR and Fl schedules for each group and each multiple schedule (regular, reversed), without the first presentation of each schedule. In the regular multiple schedule, a significant difference between the numbers of presses on the FR and FI appeared for each group except for the patients who were not given any instruction (NI group). In other words, each group except the patients NI group was able to differentiate between the two schedules during the regular multiple schedule. (3)
Table 4 Mean number of presses (SD) and p value (Wilcoxon tests
for two 'paired samples) of the difference between the mean
number of presses, on fixed-ratio and fixed-interval schedules
for each group during the multiple schedule phase

Condition  Group   FR regular      FI regular    [rho]     FR reversed
                   multiple        multiple      regular     multiple
                   schedule        schedule      multiple    schedule

Controls   WI     463.1 (70.1)   128.3 (75.6)    0.005*   406.6 (118.1)

           NI     362.0 (148.9)  184.8 (146.5)   0.016*   349.3 (180.2)

           SI     418.2 (69.8)   210.7 (107.6)   0.005*    416.6 (62.2)

Patients   WI     306.0 (78.8)   143.2 (81.2)    0.027*   255.2 (195.3)

           NI     330.8 (74.1)   280.6 (71.4)     0.224    408.4 (51.9)

           SI     261.0 (138.1)  205.2 (101.0)   0.043*   216.2 (175.9)

Condition  FI reversed     [rho]
           multiple       reversed
           schedule       multiple

Controls   199.3 (144.8)   0.036*

           115.5 (106.1)   0.006*

           188.3 (116.2)   0.005*

Patients   199.7 (101.1)    0.463

           306.8 (67.2)    0.043*

           167.8 (88.4)     0.138

FR= fixed-ratio, FI=fixed-interval. * < [rho] = 005.
First presentation of each schedule not included

A comparison of the Patients WI group and the Controls WI group numbers of presses differences between the FR and FI during the reversed multiple schedule tested the hypothesis that participants formerly presenting with delusional ideas would show less adjustment than controls when following an instruction. The difference was significant for Control WI (p=0.036), whereas it was not for Patients WI (p=0.463, see Table 4). This finding indicates that the patients did not adjust their rate of responding based on changes in environmental contingencies (i.e., the consequences of their presses), whereas the control participants did, despite the fact that patients had been able to adjust in the context of the regular multiple schedule (p.=0.027).

A second hypothesis was that participants formerly presenting with delusional ideas would also be less likely than control participants to adjust their behaviors in response to a change in environmental contingencies if they had previously formulated a self-instruction about how to respond. Results supported this hypothesis. On the reversed multiple schedule, the difference in the number of presses between the FR and FI was not significant for the Patient SI group (p= 0.138), whereas it was for the Control SI group (p=0.005, see Table 4). This result indicates that, when asked to formulate their own rule on how to behave, the controls adjusted to changes in environmental contingencies, whereas the patients did not, even though the latter had shown an adjustment to FR and FI on the regular multiple schedule ([rho] = 0.043). (4)

According to this second hypothesis, participants formerly presenting with delusional ideas should become more insensitive to changes in environmental contingencies when asked to formulate their own rule on how to behave compared to when an instruction was given to them. To test this hypothesis, the difference between FR and FI was calculated for the regular and reversed multiple schedule, respectively. This difference would indicate adjustment (i.e., a positive difference would indicate good adjustment as more presses were made on FR than FI, while a negative difference would indicate poor adjustment as FR<FI. These differences were compared between the regular and reversed multiple schedule for patients and appeared to be statistically equivalent (Mann-Whitney=13; [rho] > 0.05), which indicates no difference with respect to the contingency change when patients were given an instruction or asked to compose their own. For control participants, these differences also proved to be equivalent when comparing the WI and SI groups (Mann-Whitney=31; [rho] > 0.05), but with an adjustment to schedules changes.

When comparing control participant groups, one-way ANOVA showed that the differences between FR and FI in the regular and reversed multiple schedule proved to be statistically different across groups (F (2, 27) = 3.59, [rho] < 0.05), with LSD post-hoc comparisons showing the WI group to adjust less to contingency changes than the NI group ([rho] = 0.01), as per previous findings in the literature.

Apart from the results just reported, Wilcoxon tests for two paired samples revealed that the patients receiving no instructions did show adjustment of their responding to environmental changes when the multiple schedule was reversed (p= 0.043), despite the fact that they had not previously adjusted on the regular multiple schedule (p=0.224). When receiving no instructions, control participants adjusted on every multiple schedule, reversed or not.


The present article proposed a behavioral explanation for maintenance of delusions. In contrast to prior cognitive explanations of this phenomenon, we focused on insensitivity to environmental contingencies as a key underlying process and suggested that the verbal nature of delusions causes rule-based insensitivity to some environmental changes.

Our pilot study compared the influence of instructions on the performance of control participants and patients with schizophrenia who formerly presented with delusions in the context of a previously established behavioral pattern to be changed. We first predicted that individuals formerly presenting with delusional ideas would demonstrate more instruction-based insensitivity than non-clinical participants. The results supported this first hypothesis. Whereas control participants were able to adjust to a change in contingencies when following an instruction that initially helped them, patients were not. The study also compared the influence of given versus self-emitted instruction on sensitivity to environment changes. We predicted that insensitivity demonstrated by participants formerly presenting with delusional ideas would be greater for self-emitted instruction. This second hypothesis was not supported by the present data. A comparable lack of adjustment to environmental changes appeared for patients who were given a rule and those who devised their own.

Our results suggest that in the presence of instructions, either given or produced, participants formerly presenting with delusional ideas became more insensitive to changes in the environment than non-clinical controls. That is, when following an instruction on how to behave in a specific situation, participants formerly presenting with delusional ideas were less likely than non-clinical control participants to change their behavior according to changes in the situation.

Despite the limited number of participants of the present pilot study, our findings suggest a new understanding of the different mechanisms proposed so far to understand delusion maintenance. As regards the data-gathering bias highlighted by Garety & Freeman (1999), the formulation of a hypothesis on the functioning of the world can be conceptualized as the derivation of a rule based on data gathering. According to our results, people with schizophrenia demonstrating delusional ideas tend to be more insensitive to their environment in the presence of instructions. Such insensitivity could stop further data gathering and, as a consequence, shut off further adaptation to the environment and further rule modification as soon as a rule is formulated.

The accounts suggesting biases against disconfinnatory (Woodward et al. 2006a, b) and for confirmatory evidences (Freeman et al., 2002) can also be understood in terms of the current findings. Insensitivity to the environment observed in participants formerly presenting with delusional ideas can be viewed as an example of discarding disconfirmatoiy evidence mediated by the verbal processes of delusions. The presence of a rule formulated by patients might organize rejection or oversight of evidences contradicting the nile in a kind of self-protecting process. Consideration of these findings can also shed light on the influence of bias in favor of confirmatory evidence on the maintenance of delusions. Following the account of delusions proposed in the present article in terms of behaviors governed by self-instructions, confirmatory evidence might contribute to strengthen patients' trust in self-generated rules by validating the accuracy of the rule itself

Two main kinds of rule-governed behaviors have been identified in behavior analysis, namely tracking and pliance. Pliancc is rule-governed behavior primarily under the control of apparent speaker-mediated consequences for a correspondence between the rule and the relevant behavior. Tracking is rule-governed behavior under the control of the apparent correspondence between the rule and the way the world is arranged (Zettle & Hayes, 1982). Ontogenetically, pliance likely precedes tracking (Torncke et al. 2008). A child follows rules given by her caregivers because of the consequences those caregivers have organized for doing so. Later on, the child can rely on the direct concrete consequences of their behaviors. The very construction of rules depends itself on contact with environmental contingencies: rules can be viewed as verbal ways to apprehend the environment and need at first a precise observation of it. Maher's (1974) hypothesis of delusions as explanations of perceptual anomalies seems to parallel this direction for the formation of delusion: first observing the environment (e.g., modifications of sound perception), followed by construction of a rule (delusion). While detecting perceptual modifications, individuals with schizophrenia who formerly presented with delusional ideas might verbally try to apprehend and understand these changes, as every verbal human being would. Our results suggesting that maintenance of delusions in schizophrenia emerges from a tendency to rely too much on rules and self-instructions once established could explain why patients fail to modify their delusional ideas if it helps them to understand the perceptual anomalies they have detected.

Although these results appear promising, this pilot study comes with a number of caveats. First, these results were acquired in the context of a relatively small sample and thus require replication with larger numbers. The weak duration of this pilot study may also be responsible for a number of surprising results observed. Notably, the patients NI group failed to adjust to contingencies in the regular multiple schedule but were able to adjust during the reversed multiple schedule, suggesting that the 4 min of each schedules' presentation was too short for some participants to explore the contingencies. This reason, combined with the great diversity of adjustments of patients to environmental contingencies and sample limits, could explain why this patients NI group did not adjust their behaviors to the schedules of reinforcement during regular multiple schedule while patients in SI group did. Indeed, the instructions given two these two groups were identical during the regular multiple schedule (no instruction; SI patients were asked to complete sentences at the end of the regular multiple schedule), and these two groups should have adjusted identically during this phase of the experiment. The same reason could explain why control participants receiving instructions adjusted to changes in contingencies in the reversed multiple schedules while insensitivity has been observed in previous experiments with participants receiving instructions that finally became inaccurate. Our results show that the control participants WI group presented with less adjustment to contingencies than the NI group in the reversed multiple schedules, but did not present the specific insensitivity observed in the literature. The short duration of the regular multiple schedules may have been insufficient for participants to observe the accuracy of the instructions provided to earn as many points as possible, hence limiting their insensitivity to inversion of schedules. In other words, the 4-min duration may have been insufficient for the instructions to control their behaviors durably. When insensitivity has been observed in comparable experimental settings, the duration of the experiment before and after inversion of schedules without notification of participants was longer (32 min for example in Hayes et al., 1986; 52 min in Rosenfarb et al., 1992).

Another limitation of the present research comes from the fact that verbal capable organisms seem to have a pervasive verbal activity with whatever context confronts them. In our experiment, it was impossible to control for the presence of instructions generated by the participants, even when not asked to do so. Participants asked to generate their own instructions differed from those who received no instructions in the fact that "shaped" self-generated rules have proved to engender insensitivity to the programmed contingencies (Rosenfarb et al., 1992). But nothing prevented participants from following their own rules, even when not asked to state out loud the instructions they deducted from their previous behavior consequences

Also, the present study did not include a control for atten-tional function. However, perseveration and over-switching are frequently reported in people suffering from schizophrenia (Yogev et al. 2003), and these pathological patterns of behaviors could influence adjustment to actual contingencies. Despite the fact that the patients in the present study adapted their responding approptiately in the context of simple schedules, an evaluation of attentional function is warranted to better gauge the possible effect of this potential confounding variable. Working memory impairments are also documented in patients with schizophrenia (Lee & Park, 2005). Such impairments might also have affected adaptation during the multiple schedule since these schedules were alternated. Indeed, one certainly needs to memorize what happened previously on a schedule when confronted with this same schedule again. Thus, the evaluation of potential working memory impairments might be advisable in future replications of this research. Finally, comparing patients with and without delusional ideas would allow for investigation of the possibility that the insensitivity effect observed in the present study is linked to schizophrenia per se as opposed to the delusions.

The current pilot study suggests a number of avenues for future research. Although our results provide evidence that rule-based insensitivity is more present in people with schizophrenia who formerly presented with delusional ideas than in non-clinical participants, a more comprehensive version of the protocol needs to be provided to tease out the variables associated with this phenomenon. First, delusional ideas often involve emotional aspects not included in the current design. Furthermore, delusions are often explicitly contradicted by others, and yet the patient continues to act in accordance with them. Future research might include these variables in the current paradigm. For instance, in the presence of slightly different schedules, patients could be asked to formulate their own instruction while given a contradictory one in order to observe which one they more likely follow.

In addition, the rule-based insensitivity account suggests possible approaches to limit the deleterious effects of self-instruction on adaptation to the environment. One such approach is Acceptance Commitment Therapy (ACT, Hayes et al. 1999). ACT suggests processes of deliteralization as a mean of counteracting fusion with language and subsequent insensitivity to the environment. It also promotes a direct contact with the environment via experiential techniques to counter verbal control of behaviors. Several studies have used ACT to treat delusion and hallucinations (e.g., Bach & Hayes, 2002; Bach et al. 2012; Gaudiano & Herbert, 2006). In Bach & Hayes (2002), though patients continued to report hallucinations and delusional ideas following ACT treatment, the believability of those ideas decreased, and the rate of rehospitalization was half that of patients enrolled in TAU over a 4-month follow-up period. Future research might examine the efficacy of ACT type processes as an intervention for rule-based insensitivity in delusional populations.

Published online: 9 April 2014 Association of Behavior Analysis International 2014

DOI 10.1007/s40732-014-0029-8


Bach, P., & Hayes, S. C. (2002). The use of Acceptance and Commitment Therapy to prevent the rehospitalization of psychotic patients: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 70, 1129-1139.

Bach, P., Gaudiano, B. A., Hayes, S. C., Herbert, J. D. (2012). Acceptance and commitment therapy for psychosis: intent to treat, hospitalization outcome and mediation by believability. Psychosis, 1-9. doi: 10.1080/17522439.20 I 2.671349 Baron, A., & Galizio, M. (1983). Instructional control of human operant behavior. The Psychological Record, 33, 495-520.

Bentall, R. P. (1994). Cognitive biases and abnormal beliefs: towards a model of persecutory delusions. In A. S. David & J. Cutting (Eds.), The Neuropsychology of Schizophrenia (pp. 337-360). Hove: Lawrence Erlbaum Associates.

Catania, A. C. (1991). Glossary. In I. H. Iversen & K. A. Latta! (Eds.), Experimental analysis of behavior. Part 2 (pp. G 1--G44). Amsterdam: Elsevier.

Catania, A. C., Shimoff, E., & Matthews, B. A. (1989). An experimental analysis of rule-governed behavior. In S. C. Hayes (Ed.), Rule-governed behavior. Cognition, contingencies, and instructional control (pp. 119-150). New-York: Plenum Press.

Catania, A. C., Shimoff, E. H., & Matthews, B. A. (1990). Properties of rule-governed behaviour and their implications. In D. E. Blackman & H. Lejeune (Eds.), Behaviour analysis in theory and practice (pp. 215-230). Hove & London: Erlbaum.

Corcoran, R., Cahill, C., & Frith, C. D. (1997). The appreciation of visual jokes in people with schizophrenia: A study of imentalizing ability. Schizophrenia Research, 24, 319-327.

Corcoran, R., Mercer, G., & Frith, C. D. (1995). Schizophrenia, symptomatology and social inference: investigating "theory of mind" in people with schizophrenia. Schizophrenia Research, 17, 5-13.

Doody, G. A., Gotz, M., Johnstone, E. C., Frith, C. D., & Owens, D. G. (1998). Theory of mind and psychoses. Psychological Medicine, 28, 397-405.

Dudley, R. E. J., John, C. H., Young, A. W., & Over, D. E. (1997). Normal and abnormal reasoning in people with delusions. British Journal of Clinical Psychology, 36, 243-258.

Ferster, C. B., & Skinner, B. F. (1957). Schedules of reinforcement. New York: Appleton-Century-Crofts.

Freeman, D., Garety, P. A., & Kuipers, E. (2001). Persectitoty delusions: developing the understanding of belief maintenance and emotional distress. Psychological Medicine, 31, 1293-1306.

Freeman, D., Garety, P. A., Kuipeis, E., Fowler, D., & Bebbington, P. E. (2002). A cognitive model of persecutory delusions. British Journal of Clinical Psychology, 41, 331-347.

Freeman, D., Garety, P. A., Kuipers, E., Fowler, D., Bebbington, P. E., & Dunn, G. (2007). Acting on persecutory delusions: the importance of safety seeking. Behaviour Research and Therapy 45, 89-99.

Frith, C. D., & Corcoran, R. (1996). Exploring 'theory of mind' in people with schizophrenia. Psychological Medicine, 26, 521-530.

Gaudiano, B. A., & Herbert, J. D. (2006). Acute treatment of inpatients with psychotic symptoms using Acceptance and Commitment Therapy: Pilot results. Behaviour Research and Therapy, 44, 415-437.

Garety, P. A., & Freeman, D. (1999). Cognitive approaches to delusions: a critical review of theories and evidence. British Journal of Clinical Psychology, 38, 113-154.

Garety, P. A., Hemsley, D. R., & Wessely, S. (1991). Reasoning in deluded schizophrenic and paranoid patients: biases in performance on a probabilistic inference task. Journal of Nervous and Mental Disease, 179, 194-201.

Hayes, S. C., Brownstein, A. J., Zettle, R. D., Rosenfarb, I. S., & Korn, Z. (1986). Rule-governed behavior and sensitivity to changing consequences of responding. Journal of the Experimental Analysis of Behavioz; 45, 237-256.

Hayes, S. C., Strosahl, K., & Wilson, K. G. (1999). Acceptance and Comminnent Therapy: An experiential approach to behavior change. New York: Guilford Press.

Hemsley, D. R., & Garety, P. A. (1986). Formation and maintenance of delusions: a Bayesian analysis. British Journal of Psychiatry 149, 51-56.

Huq, S. F., Garety, P. A., & Hemsley, D. R. (1988). Probabilistic judgements in deluded and non-deluded subjects. Quarterly Journal of Experimental Psychologv 40, 801-812.

Kaney, S., & Bentall, R. P. (1989). Persecutory delusions and attributional style. British Journal of Medical Psychology, 62,191-198.

Kinderman, P., & Bentall, R. P. (1996). Self-discrepancies and persecutory delusions: evidence for a model of paranoid ideation. Journal of Abnormal Psychology, 105, 106-113.

Kinderman, P., & Bentall, R. P. (1997). Causal attributions in paranoia and depression: internal, personal, and situational attributions for negative events. Journal of Abnornzal Psychology, 106, 341-345.

Kinderrnan, R, & Bentall, R. P. (2000). Self-discrepancies and causal attributions: studies ofhypothesized relationships. British Journal of Clinical Psychology, 39, 255-273.

Lee, J., & Park, S. (2005). Working memory impairments in schizophrenia: a meta-analysis. Journal ofAbnormal Psychology, //4(4), 599611. doi: 10.1037/0021-843X.114.4.599.

Lyon, H. M., Kancy, S., & Bentall, R. P. (1994). The defensive function of persecutory delusions: evidence from attribution tasks. British Journal of Psychiatry 164, 637-646.

Maher, B. A. (1974). Delusional thinking and perceptual disorder. Journal of Individual Psychology, 30, 98-1 13.

Maher, B. A. (1988). Anomalous experience and delusional thinking: The logic of explanations. In T. F. Oltmanns & B. A. Maher (Eds.), Delusional Beliefs (pp. 15-33). New York: Wiley.

Maher, B. A. (1999). Anomalous experience in everyday life: its significance for psychopathology. The Monist, 82, 547-570.

Moritz, S., & Woodward, T. S. (2006). A generalized bias against disconfirmatory evidence in schizophrenia. Psychiatry Research, 142,157-165.

Rosenfarb, I. S., Newland, M. C., Brannon, S. E., & Howey, D. S. (1992). Effects of self-generated rules on the development of schedule-controlled behavior. Journal of the Experimental Analysis of Behavior 58,1 07 1 2 1 .

Sarfati, Y. (2000). Deficit of the theory-of-mind in schizophrenia: clinical concept and review of experimental arguments. Canadian Journal of Psychiatry, 45, 363-368.

Sharp, H. M., Fear, C. F., & Healy, D. (1997). Attributional style and delusions: an investigation based on delusional content. European Psychiatry, 12, 1-7.

Shimoff, E., & Catania, A. C. (1998). The verbal governance o f behavior. In K.. A. Latta', M. Perone, & I. H. Iversen (Eds.), Handbook of human operant behavioral research methods. New York: Plenum Press.

Startup, H., Freeman, D., & Garety, P. A. (2007). Persecutory delusions and catastrophic worry in psychosis: developing the understanding of delusion distress and persistence. Behaviour Research and Therapy 45, 523-527.

Tomeke, N., Luciano, C., & Valdivia Salas, S. (2008). Rule-Governed Behavior and Psychological Problems. International Journal of Psychology and Psychological Therapy, 8,141-156.

Vaughan, M. (1989). Rule-governed behavior in behavior analysis. A theoretical and experimental history. In S. C. Hayes (Ed.), Rule-governed behavior Cognition, contingencies, and instructional control (pp. 97-118). New-York: Plenum Press.

Woodward, T. S., Moritz, S., & Chen, E. Y. H. (2006a). The contribution of a cognitive bias against disconfirmatory evidence (BADE) to delusions: a study in an Asian sample with first episode schizophrenia spectrum disorders. Schizophrenia Research, 83, 297-298.

Woodward, T. S., Moritz, S., Cuttler, C. C., & Whitman, J. (2004). A generalized cognitive deficit in integrating disconfirmatory evidence underlies delusion maintenance in schizophrenia. Schizophrenia Research, 67, 79.

Woodward, T. S., Moritz, S., Cuttler, C. C., & Whitman, J. (2006b). The Contribution of a Cognitive Bias Against Disconfirmatoiy Evidence (BADE) to Delusions in Schizophrenia. Journal of Clinical and Experimental Neuropsychology, 28, 605-617.

Wulfert, E., Greenway, D. E., Farkas, P., Hayes, S. C., & Dougher, M. J. (1994). Correlation between self-reported rigidity and rule-governed insensitivity to operant contingencies. Journal of Applied Behavior Analysis, 27, 659-671.

Yogev, H., Hadar, U., Gutman, Y., & Sirota, P. (2003). Perseveration and over-switching in schizophrenia. Schizophrenia Research, 6/(2-3), 315-321.

Zettle, R. D., & Hayes, S. C. (1982). Rule governed behavior: A potential theoretical framework for cognitive behavior therapy. In P. C. Kendall (Ed.), Advances in cognitive behavioral research and therapy (pp. 73-118). New York: Academic Press.

(1) Accepted by CCPPRB of Picardie under reference 03H19.

(2) Fixed reinforcement schedules have been preferred to variable ones so that adjusted patterns of response are distinguished clearly across the two keys of responses, and lack of adjustment is easily observed during the reversed part of the experiment.

(3.) Keeping the data from the first presentation of each schedule in the data set revealed that patients in SI groups did not adjust to contingencies changes in regular phase, as controls WI and patients NI groups did not. In sum, the changes comparing the results without the first presentation of the results are from adjustment to non-adjustment to the programmed contingencies. Presentation of each schedule engenders too eratic behaviors since participants spend this first presentation exploring the contingencies.

(4.) As it turned out, the instructions formulated by control participants and patients corresponded closely in all but one case to those actually given to the participants in the with-instruction groups. Only one patient formulated slightly different instructions and, even in this case, these instructions, when followed, engendered similar patterns of response This participant thought that in the ratio program, he had to press strongly among a sequence of light presses, while in the interval program, three presses followed by a pause were needed to earn a point. Following these instructions, he would have emitted more responses on the ratio program, exactly as with a more accurate instruction.

J. L. Monestes

Mental Health Service, University of Reunion, Reunion, France

M. Villatte

Evidence Based Practice Institute, Seattle, USA


I. Stewart

National University of Ireland Galway, Galway, Ireland


G. Loas

Cliniques Universitaires de Bruxelles, Hopital Erasme,

Universite libre de Bruxelles (ULB), Brussel, Belgium


J. L. Monestes (*)

Pole de Sante Mentale, CHU de la Reunion,

BP 350, 97448 Saint-Pierre Cedex, France


J. L. Monestes

Laboratoire Epsylon EA 4556, Montpellier, France
COPYRIGHT 2014 The Psychological Record
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Monestes, J.L.; M. Villatte; Stewart, I.; Loas, G.
Publication:The Psychological Record
Article Type:Report
Date:Jun 1, 2014
Previous Article:The influence of native acquisition of Chinese on mental rotation strategy preference: an eeg investigation.
Next Article:Motivation Behavior Analysis A Critique.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters