The importance of individual differences in intelligence and working memory capacity in predicting the ability to intentionally suppress thoughts was investigated. Sixty participants completed a thought suppression task, and measures of working memory capacity (OSPAN), fluid intelligence (Raven’s Matrices), and crystallised intelligence (the National Adult Reading Test). As predicted, the results indicated that more effective thought suppression was independently related to higher working memory capacity and greater fluid intelligence, but was unrelated to crystallised intelligence. The findings have theoretical implications for understanding the mechanisms underlying a failure to inhibit unwanted intrusions and clinical implications for disorders involving high levels of intrusive thoughts and memories.
Research in the past two decades has shown that unpleasant, intrusive thoughts and memories are common everyday phenomena that tend to be present at an abnormally high frequency in clinical populations, and may be implicated in cases where the condition has a more chronic course (Brewin, Reynolds, & Tata, 1998). Attempts to avoid or suppress these intrusive cognitions are ubiquitous both across the anxiety disorders and depression (e.g. Brewin, 1998, Purdon and Clark, 1993), and may indicate a breakdown in normal inhibitory processes designed to exclude irrelevant material from entering consciousness (Brewin, 2001). However, deliberate attempts at thought suppression can be counter-productive, helping instead to return unwanted thoughts to mind and to reinforce the very state of mind an individual is trying to avoid (Wenzlaff & Wegner, 2000). In this study we test the hypothesis that the success with which someone is able to suppress an unwanted thought is related to individual differences in certain kinds of mental ability, specifically fluid intelligence and working memory capacity.
In a well-known laboratory study Wegner, Schneider, Carter, and White (1987) found evidence for the inadequacy of attempts at deliberate thought suppression. Participants were instructed not to think about a white bear and to indicate the thought’s return with verbal report or bell-rings. The results showed that during a five-minute suppression period participants, on average, thought about a white bear more than once per minute. Moreover, there was evidence of thought rebound, showing that prior suppression can significantly enhance the degree of subsequent expression once the thought suppression instructions had been lifted. Others have since used identical or similar protocols with somewhat inconsistent results. While the majority have found evidence of thought suppression or post-suppression rebound (see Abramowitz, Tolin and Street, 2001, Wenzlaff and Wegner, 2000, for a review), several have failed to do so (e.g. Merckelbach, Muris, van den Hout and de Jong, 1991, Muris, Merckelbach, van den Hout and de Jong, 199). In two studies that only partly supported the existence of a post-suppression rebound, Rutledge, Hollenberg, and Hancock (1993) found that those participants who obtained lower scores on a scholastic aptitude test were more likely to experience difficulty in suppressing unwanted thoughts. Their results suggest that varying success at thought suppression may be explained by individual differences in mental abilities.
Intelligence is often conceptualised in terms of speed of information-processing, the quality or quantity of information represented, executive processes and/or processing capacity. Yet intelligent behaviour depends on the ability to suppress task-irrelevant information as well as the ability to activate task-relevant information, as individual and group differences in performance on interference-sensitive tasks suggests (Dempster, 1991). For example, research comparing good and poor comprehenders has confirmed that the ability to maintain attention only to relevant information in the presence of irrelevant peripheral and contradictory stimuli is important to comprehension (Gernsbacher & Faust, 1991). Less skilled comprehenders reject less efficiently the inappropriate meanings of ambiguous words, the incorrect forms of homophones, and the highly typical but absent members of scenes. Possibly related to these observations is the well-established finding that low intelligence is a risk factor for posttraumatic stress disorder (PTSD), a condition characterised by high levels of memories that intrude into consciousness against the person’s will and interfere with concentration and memory (Brewin, Andrews, & Valentine, 2000). This holds true even when intelligence is measured prior to trauma exposure (Macklin, Metzger, Litz, McNally, Lasko et al., 1998).
Rosen and Engle (1998) reported that a higher working memory capacity was related to having fewer irrelevant intrusions in a task involving the learning of successive word lists. Rosen and Engle attempted to measure working memory capacity using a dual task intended to tap participants’ ability to sustain, divide and switch attention between processing an arithmetic string and learning a sequence of words. Engle, Cantor, and Carullo (1997) previously argued that individual differences in working memory capacity reflect differential abilities to bring domain-free, focussed attention to bear on cognitive tasks. As such, measures of working memory capacity were assumed to be valid measures of constructs such as the central executive (Baddeley & Hitch, 1974) and the supervisory attentional system (Norman & Shallice, 1986). Rosen & Engle (1998) further proposed that working memory capacity reflects the capacity for controlled, sustained attention in the face of interference or distraction.
A strong relationship between working memory capacity and fluid intelligence has been found, suggesting that both constructs reflect the capacity to keep mental representations active (Engle, Tuholski, Laughlin, & Conway, 1999b). Following Horn and Cattell (1967), fluid intelligence (gF) refers to the ability to solve novel problems and adapt to new situations. It is thought to be nonverbal and relatively free from the influence of culture, in contrast to crystallised intelligence (gC) which refers to acquired skills and knowledge and is dependent on education and cultural background. In the experiment by Engle et al. (1999b) no relationship was found between gC and working memory capacity. Kyllonen and Christal (1990) examined the relationship between working memory capacity and reasoning ability (a central factor of gF) in order to ascertain whether working memory capacity is the psychological mechanism responsible for gF. They found that correlations between working memory capacity and the reasoning factor ranged from 0.80 to 0.90.
The hypothesis tested here is that success at deliberate thought suppression will be associated with individual differences in fluid intelligence and working memory capacity, but not with crystallised intelligence. We further examined whether fluid intelligence and working memory capacity make separate or overlapping contributions to predicting success at thought suppression.