Thought Suppression and Psychopathology: The White Gedeer Returns

I am pleased to introduce Dr. Nic Hooper, a Lecturer in Psychology at the University of the West of England (Bristol). Nic’s research in Relational Frame Theory and Acceptance and Commitment is wide and varied but always with a practical focus engaged on meaningful behavioral change. His publications include work in health psychology (e.g., smoking cessation; obesity), clinical psychology, and educational psychology. Here Nic describes some very interesting recent research work along with his colleagues Louise McHugh and Ian Stewart on behavioral modelling of thought suppression effects in psychopathology from a relational frame theory perspective. Enjoy – Dr. Ian Tyndall, University of Chichester, UK.

We have all wanted to get rid of an unwanted thought at some time or another. However, not only does empirical evidence suggest that thought suppression is difficult to achieve (Wegner, Schneider, Carter, & White, 1987) but it may play a role in the development and maintenance of psychopathology (Wenzlaff & Wegner, 2000). One way Wegner (1989) accounted for why thought suppression attempts may be difficult was with the ‘Environmental Cueing Hypothesis’. He suggested that when trying to suppress an unwanted thought, we will often think about something else instead i.e., a distractor. However, through consistent use, the distractor itself will become related to, and thus, ironically, a cue, for the unwanted thought. Furthermore, the more distractors we use, the more potential cues, and thus the more difficult it becomes to successfully suppress. Various empirical studies have supported this explanation (Najmi & Wegner, 2008; Wegner & Erber, 1992); however, in the past decade, Stimulus Equivalence and Relational Frame Theory (RFT) research have provided a noteworthy addition to our understanding in this area, and describing that contribution is the purpose of this blog.

Murray Sidman demonstrated that if a participant was taught that A is the same as B, and that B is the same as C then a number of ‘untaught’ relations will emerge (note: three stimuli are used in the current example for ease of understanding but theoretically it is possible that many stimuli may be trained to be functionally equivalent). Specifically, participants would be able to reverse (B is the same as A, C is the same as B) and combine (A is the same as C, C is the same as A) the relations. Hayes, Barnes-Holmes and Roche (2001) extended this idea further to suggest that ‘derived relational responding’ does not happen just in accordance with equivalence or sameness relations but in accordance with a host of other relations (or frames) also including, for example, opposition, comparison, distinction, hierarchy, deixis, etc. Furthermore, they suggested that in the case of all such relational frames, three types of derivation can be seen, namely, mutual entailment, combinatorial entailment and transformation of function. Mutual entailment describes how, when a relation between two stimuli is taught (e.g., A is more than B), a second relation in the opposite direction is derived (i.e., B is less than A). Combinatorial entailment describes the combination of two or more trained relations. For example, if taught that ‘A is the same as B’, and ‘B is the same as C’, we can combine those relations to derive the untrained relations ‘A is the same as C’ and ‘C is the same as A’. Finally, transformation of function (ToF) refers to the way that the psychological functions of a stimulus in a derived relation with another stimulus can change depending on the functions of the latter and the nature of the derived relation between the two. For example, in Whelan and Barnes-Holmes (2004), one arbitrary stimulus was established as a conditioned punisher and subsequently other stimuli derived as being the same as it became punishing while stimuli derived as the opposite of it became reinforcing.

Although some of this technical language can sometimes be intimidating, the implication that derived learning has for thought suppression is simple; if a suppression function is acquired by a target stimulus (i.e., so that the person tries to suppress that stimulus) then the function of other stimuli in both directly trained and derived relations with it may change accordingly without further training being needed. In lay terms, thought suppression would be difficult not just because of stimuli directly associated with the target stimulus, as Wegner suggested, but because stimuli in derived relations would also interfere with the suppression attempt.

Hooper, Saunders and McHugh (2010) provided an empirical demonstration of this phenomenon. In the study, participants were induced to derive three equivalence relations between textual stimuli, which can be represented using the alphanumerics A1-B1-C1, A2-B2-C2, A3-B3-C3. This was done for each of the three relations by training participants directly that the A and B stimuli were the same and that the B and C stimuli were the same and then testing for the derivation of sameness between A and C. The actual stimuli used were a mixture of natural language words and nonsense words. For example, the first stimulus class (A1-B1-C1) included the three words ‘Bear’ (A1), ‘Boceem’ (B1) and ‘Gedeer’ (C1). Hence, in this case, based on the trained sameness relations BearBoceem and BoceemGedeer, a combinatorially entailed relation of sameness between ‘Bear’ and ‘Gedeer’ can be derived and indeed participants were shown to derive this as well as the two combinatorially entailed relations predicted for the other two stimulus sets (i.e., A2-C2 and A3-C3) in a subsequent non-reinforced testing phase.

Upon successfully completing the equivalence training and testing phase, participants were required to complete a suppression procedure involving two tasks. First they had to suppress all thoughts of the word ‘bear’, which was the A1 stimulus in the first equivalence relation discussed above i.e. the word ‘bear’ was given a suppression function. Second, participants had to watch a computer screen on which different words would appear for 10-second time intervals. Amongst the words being shown was the target word (‘bear’), the word directly trained to it (i.e., B1 – ‘boceem’) and the word that participants had derived as being the same as the target (i.e., C1 – ‘gedeer’), as well as the words from the other two equivalence classes and random control words. Participants were told that they were in control of the program, so that if they wanted to remove any word from the screen then they could do so by pressing the space bar. It was expected that participants would remove the word ‘bear’ from the screen given that they had been asked to suppress this word. Of more interest was whether they might remove any other words and in particular whether they would remove either the directly trained or derived words. If this were the case then it would suggest that the suppression functions attached to the word ‘bear’ had transferred to these words.

The results showed that this was the case; participants tended to remove not just the word ‘bear’ but also the words ‘boceem’ and ‘gedeer’ from the screen to a significantly greater extent than other words. The finding that ‘boceem’ was removed echoed the theory of Wegner and colleagues who had argued that stimuli (e.g., distractors) directly related to a target may subsequently interfere with attempted suppression. However, the finding that participants also removed the derived word ‘gedeer’ was of particular note since this demonstrated that thought suppression is difficult not just based on interference by directly related stimuli but also because it can be affected by indirect or derived relations.

This effect has since been replicated (Galera-Barbero et al., 2012) and extended to the relational frame of opposition (Stewart, et al., 2015). It has also been used to show that people will avoid acting in previously chosen ways in an attempt to avoid target, trained and derived stimuli (Hooper, Stewart, Duffy, Freegard, & McHugh, 2012). Put together, the studies tell an interesting story: the generative nature of language makes thought suppression attempts even more difficult than Wegner first imagined, and people will often change their overt behavior in the service of avoidance. At this point, there is a chance that you might be asking yourself the question ‘why are these studies important?’ This brings us right back to the beginning of this blog when we referred to psychopathology. These studies are important because they hold valuable information for therapeutic intervention – in particular, that such interventions should attempt to undermine avoidance as a way to manage unwanted thoughts.