Introduction
Over the past decade there has been notable progress in understanding the cognitive mechanisms underlying a number of individual psychotic experiences. The greatest attention has been on delusional beliefs. A reasoning process repeatedly linked to delusions has been the presence of jumping to conclusions (JTC). This data-gathering bias has been found in 30 out of 38 clinical studies (see Garety & Freeman, Reference Garety and Freeman1999, Reference Garety and Freeman2013). It is hypothesized that a limited data search leads to a rapid acceptance of implausible ideas, without consideration of alternative explanations. Initial evidence indicates that JTC may be a predictor of delusion persistence (Menon et al. Reference Menon, Mizrahi and Kapur2008; Dudley et al. Reference Dudley, Daley, Nicholson, Shaftoe, Spencer, Cavanagh and Freeston2013), and, importantly, that reducing the bias may reduce the delusion (Moritz et al. Reference Moritz, Vitzthum, Randjbar, Veckenstedt and Woodward2010; Waller et al. Reference Waller, Freeman, Jolley, Dunn and Garety2011). But what are the cognitive factors that contribute to the occurrence of JTC? In this study we test two hypotheses. First, JTC may result from information-processing limitations, as marked by working memory performance. Second, JTC may be a motivated strategy used to reduce difficulties coping with uncertainty.
JTC is assessed using ‘the beads task’, an experimental test originally based upon Bayesian probabilistic inference (Garety et al. Reference Garety, Hemsley and Wessely1991; Dudley et al. Reference Dudley, John, Young and Over1997; Warman et al. Reference Warman, Lysaker, Martin, Davis and Haudenschield2007; Lincoln et al. Reference Lincoln, Ziegler, Mehl and Rief2010). Participants are shown two jars varying in the proportion of differently coloured beads (e.g. one jar has 60 black beads and 40 yellow beads, the other jar has 40 black beads and 60 yellow beads). The jars are then hidden from view and the participant is told that the experimenter has selected one jar. The participant is asked to decide with certainty which jar has been chosen (e.g. the mainly black bead jar or the mainly yellow bead jar). The participant can request to see as many beads, drawn one by one from the selected jar, as he or she would like before making a decision. JTC is defined as deciding after being shown two or fewer beads. There is an easy version of the task, where the difference in the proportions of the two colours of bead is large (80:15), and a more difficult version of the task, where the two colours are present in proportions that are almost equal (60:40). About half of patients with delusions show JTC on the 80:15 beads task and about one-third on the 60:40 version (Garety & Freeman, Reference Garety and Freeman2013).
Two recent large studies have tested the association of JTC with neuropsychological functioning. Although neuropsychological problems have been linked to the negative rather than the positive symptoms of psychosis (e.g. Ventura et al. Reference Ventura, Hellemann, Thames, Koellner and Nuechterlein2009), there is a plausible case for expecting a contribution of working memory to the beads task – task instructions, two alternative possibilities, and the previous beads drawn have to be kept in mind. Garety et al. (Reference Garety, Joyce, Jolley, Emsley, Waller, Kuipers, Bebbington, Fowler, Dunn and Freeman2013) tested 126 patients with delusions, comparing working memory and pre-morbid intellectual functioning in those patients who jumped to conclusions with those who did not. There were no differences between the two groups in pre-morbid intelligence quotient (IQ) estimated with the Wechsler Test of Adult Reading (Wechsler, Reference Wechsler2001). But on the 60:40 beads task in particular, there were associations of JTC with poorer working memory performance. The effect sizes were small to moderate. Falcone (Reference Falcone2013) tested 112 patients with first-episode psychosis. IQ was assessed with five subtests of the working memory tasks from the Wechsler Adult Intelligence Scale III (WAIS-III; Wechsler, Reference Wechsler1997). There were strong associations of JTC with IQ and working memory. The results of these two large studies are consistent with a number of earlier initial reports (e.g. Woodward et al. Reference Woodward, Mizrahi, Menon and Christensen2008), including a study indicating that data gathering may improve when memory load in the beads task is reduced (Menon et al. Reference Menon, Pomarol-Clotet, McKenna and McCarthy2006). However, a study with 29 patients with early psychosis did not find a significant association of JTC with verbal working memory (Ormrod et al. Reference Ormrod, Shaftoe, Cavanagh, Freeston, Turkington, Price and Dudley2012). Given the results of the largest studies, there are increasing grounds to believe that working memory limitations contribute to the presence of JTC.
An alternative perspective is that JTC is a motivated top-down strategy to forestall the presence of distressing uncertainty. Rather than dealing with doubt, and therefore potential confusion, a decision is rapidly made. A meta-cognitive belief – that uncertainty is to be avoided – could guide the data-gathering process when a decision is made. A potentially useful concept here is ‘intolerance of uncertainty’, developed to understand generalized anxiety disorder (Dugas et al. Reference Dugas, Gagnon, Ladouceur and Freeston1998). Intolerance of uncertainty is hypothesized to lead to magnification of problems in situations of little realistic risk, and worry is a person's misguided attempt to regain control over these magnified threats. The Intolerance of Uncertainty Scale (IUS) was developed to assess the concept (Freeston et al. Reference Freeston, Rhéaume, Letarte, Dugas and Ladouceur1994). Consistent with the idea that JTC is a motivated strategy to reduce the occurrence of distressing uncertainty, Broome et al. (Reference Broome, Johns, Valli, Woolley, Tabraham, Brett, Valmaggia, Peters, Garety and McGuire2007) found higher levels of intolerance of uncertainty to be associated with less data gathering in the beads task in a combined high risk of psychosis (n = 35) and non-clinical control (n = 23) group.
Nonetheless, the relationship of data gathering with intolerance of uncertainly may not be straightforward. In their original scale development report, Freeston et al. (Reference Freeston, Rhéaume, Letarte, Dugas and Ladouceur1994) speculated a potentially reverse effect: that intolerance of uncertainty may lead to ‘high degrees of evidence before a decision can be made’. Furthermore, Dudley et al. (Reference Dudley, Shaftoe, Cavanagh, Spencer, Ormrod, Turkington and Freeston2011) found no association between JTC and intolerance of uncertainty in 77 patients recruited from a first-episode service. Similarly, in individuals with eating disorders, Sternheim et al. (Reference Sternheim, Startup and Schmidt2011) found no association of intolerance of uncertainty with performance on the beads task. It is also notable that a concept related to intolerance of uncertainty is ‘need for closure’, defined as ‘individuals’ desire for a firm answer to a question and an aversion toward ambiguity’ (Kruglanski & Webster, Reference Kruglanski and Webster1996). However, in a study with 187 patients with psychosis (Freeman et al. Reference Freeman, Garety, Kuipers, Colbert, Jolley, Fowler, Dunn and Bebbington2006), need for closure was unrelated to JTC. Although based upon similar concepts, the standard measures of intolerance of uncertainty and need for closure show only small to moderate associations (Berenbaum et al. Reference Berenbaum, Bredemeier and Thompson2008). Intolerance of uncertainty, and its relationship to JTC, remains to be tested in a large group of patients with current delusions.
The aim of the present study was to examine JTC in patients with delusions in relation to both neuropsychological functioning and intolerance of uncertainty. It was predicted that JTC would be associated with poorer working memory, lower IQ and greater intolerance of uncertainty. The association of JTC with working memory was expected to remain, even when controlling for levels of intellectual functioning. The association of JTC with intolerance of uncertainty was expected to remain, even when controlling for levels of worry. Hence information-processing constraints and top-down motivations were both expected to contribute to the occurrence of JTC.
Method
Participants
A total of 123 patients with persecutory delusions completed the JTC task during the baseline assessment (prior to randomization) of a clinical trial (ISRCTN23197625) (Freeman et al. Reference Freeman, Dunn, Startup and Kingdon2012 a). Patients with persecutory delusions were recruited from two mental health National Health Service (NHS) Trusts: Oxford Health NHS Foundation Trust and Southern Health NHS Foundation Trust. The inclusion criteria were: a current persecutory delusion as defined by Freeman & Garety (Reference Freeman and Garety2000); scoring at least 3 on the conviction scale of the Psychotic Symptom Rating Scales (PSYRATS)-delusions scale (i.e. at least 50% conviction in the delusion) (Haddock et al. Reference Haddock, McCarron, Tarrier and Faragher1999); that the delusion had persisted for at least 3 months; a diagnosis from the treating clinical team of ‘non-affective psychosis’, including schizophrenia, schizo-affective disorder and delusional disorder; a clinically significant level of worry, as indicated by scores above 44 on the Penn State Worry Questionnaire (PSWQ; Startup & Erickson, Reference Startup, Erickson, Davey and Wells2006); aged between 18 and 65 years; and no changes to medication in the past month. Only eight patients with persecutory delusions were excluded from entering the trial because of an insufficient level of worry as assessed by the PSWQ. Criteria for exclusion were: a primary diagnosis of alcohol or substance dependency or personality disorder; organic syndrome or learning disability; a command of spoken English inadequate for engaging in therapy; and currently having individual cognitive–behavioural therapy.
Assessments
JTC: the beads task (Dudley et al. Reference Dudley, John, Young and Over1997; Garety et al. Reference Garety, Freeman, Jolley, Dunn, Bebbington, Fowler, Kuipers and Dudley2005)
Data gathering was assessed with a probabilistic reasoning task that has been extensively used with individuals with delusions. Participants are asked to request as many pieces of evidence (coloured beads) as they would like before making a decision (from which of two hidden jars the beads are drawn). The two jars have beads of two different colours in opposite ratios. The ratio of beads used in the current study was 60:40. The key variable is the number of beads requested before making a decision. Requesting two or fewer beads is classified as JTC. A categorical method of assessing JTC is used to capture the extreme form of the bias and because the number of draws to decision is not a normally distributed continuous scale (the informational value of each additional bead varies).
WAIS-III (Wechsler, Reference Wechsler1997)
Three working memory tasks were used: digit span forwards (repeating back series of numbers), digit span backwards (repeating back series of numbers in reverse), and letter–number sequencing (sorting and recalling a series of letters and numbers). The demands on working memory increase across each of these three tasks. These are the most commonly used working memory tasks in schizophrenia research (Nuechterlein et al. Reference Nuechterlein, Barch, Gold, Goldberg, Green and Heaton2004).
Wechsler Abbreviated Scale of Intelligence (WASI)
The WASI (Wechsler, Reference Wechsler1999) is a standardized short and reliable measure of intelligence. The vocabulary and matrix reasoning subtests were used to obtain an estimate of IQ.
IUS
The IUS (Freeston et al. Reference Freeston, Rhéaume, Letarte, Dugas and Ladouceur1994) has 27 items each rated on a 1 to 5 scale. Higher scores indicate greater levels of intolerance of uncertainty. There are four subscales derived from factor analysis (Berenbaum et al. Reference Berenbaum, Bredemeier and Thompson2008): desire for predictability (e.g. ‘Unforeseen events upset me greatly’); a tendency to become paralysed in the face of uncertainty (e.g. ‘When I am uncertain I can't function very well’); distress in the face of uncertainty (e.g. ‘The ambiguities in life stress me’); and inflexible uncertainty beliefs (e.g. ‘Being uncertain means that a person is disorganized’). These subscales in the present study had high levels of internal reliability: desire for predictability, Cronbach's α = 0.85; uncertainty paralysis, Cronbach's α = 0.85; uncertainty distress, Cronbach's α = 0.80; inflexible certainty beliefs, Cronbach's α = 0.76.
PSYRATS-delusions
The PSYRATS-delusions (Haddock et al. Reference Haddock, McCarron, Tarrier and Faragher1999) is a six-item multidimensional measure. It assesses the conviction, preoccupation, distress and disruption associated with delusions. Symptoms over the last week are rated. Higher scores indicate greater severity. The two assessors in the current study had high inter-rater reliability (n = 20, intra-class correlation coefficient = 0.99).
Positive and Negative Syndrome Scale (PANSS)
The PANSS (Kay, Reference Kay1991) is a 30-item rating instrument developed for the assessment of patients with schizophrenia. Symptoms over the last week were rated (i.e. currently present). Higher scores indicate the greater presence of psychiatric symptoms. The two assessors in the current study had high inter-rater reliability (n = 20, PANSS total intra-class correlation coefficient = 0.91).
Green et al. Paranoid Thoughts Scale (GPTS)
The GPTS (Green et al. Reference Green, Freeman, Kuipers, Bebbington, Fowler, Dunn and Garety2008) is a 32-item measure of paranoid thinking. Part A assesses ideas of reference (e.g. ‘It was hard to stop thinking about people talking about me behind my back’) and part B assesses ideas of persecution (e.g. ‘I was convinced there was a conspiracy against me’). Each item is rated on a five-point scale. Higher scores indicate greater levels of paranoid thinking. The scale was completed for the period of the previous fortnight. The internal reliability of the scale was high (Cronbach's α for part A was 0.90, and for part B it was 0.94).
PSWQ
The PSWQ (Meyer et al. Reference Meyer, Miller, Metzger and Borkovec1990) is the most established measure of trait worry style and has been used in non-clinical and clinical populations (for a review, see Startup & Erickson, Reference Startup, Erickson, Davey and Wells2006). Each of the 16 items is rated on a five-point scale. Higher scores indicate a greater tendency to worry. In the current study Cronbach's α for the scale was 0.79.
Perseverative Thinking Questionnaire (PTQ)
The PTQ (Ehring et al. Reference Ehring, Zetsche, Weidacker, Wahl, Schönfeld and Ehlers2011) is a 15-item questionnaire asking about how a person typically thinks about negative problems (e.g. ‘The same thoughts keep going through my mind again and again’), with each item assessed on a 0 to 4 scale. Higher scores indicate greater levels of repetitive negative thinking. Cronbach's α for the scale was 0.92.
Analysis
Analyses were carried out using SPSS (version 20.0, release 20.0.0; IBM, USA). The patient group was divided into two: those showing JTC and those who did not. To test the two primary hypotheses the groups were then compared using t tests on the measures of working memory and intolerance of uncertainty. Binary logistic regressions with JTC as the dependent variable (0 = no JTC, 1 = JTC) were then carried out to allow for analyses with covariates. All hypothesis testing was two-tailed.
Results
Demographics
The demographic details of the participants are summarized in Table 1. Levels of persecutory delusion were high as assessed using the PSYRATS and the GPTS, with scores similar to those in other samples selected for current persecutory delusions (e.g. Foster et al. Reference Foster, Startup, Potts and Freeman2010). Hallucinations were common, with 75 of the patients scoring four or above on the PANSS hallucination item. Only seven patients scored four or above on the PANSS grandiosity item.
Table 1. Basic demographic and clinical information
s.d., Standard deviation; NOS, not otherwise specified; PSYRATS, Psychotic Symptom Rating Scales; PANSS, Positive and Negative Syndrome Scale; GPTS, Green et al. Paranoid Thoughts Scale; IQ, intelligence quotient.
JTC
The mean number of beads drawn on the 60:40 task was 7.6 (s.d. = 5.9). This is comparable with that found in the study of Garety et al. (Reference Garety, Joyce, Jolley, Emsley, Waller, Kuipers, Bebbington, Fowler, Dunn and Freeman2013) who reported a mean of 7.2 (s.d. = 5.7) (n = 126). In the current study 30 patients (24.4%) with persecutory delusions jumped to conclusions. Again, this rate of JTC is comparable with that found in the study of Garety et al. (Reference Garety, Joyce, Jolley, Emsley, Waller, Kuipers, Bebbington, Fowler, Dunn and Freeman2013) (27% showing JTC), but it is lower than that reported by So et al. (Reference So, Freeman, Dunn, Kapur, Kuipers, Bebbington, Fowler and Garety2012) (40%) (n = 184). There were no differences between those who jumped to conclusions and those who did not in levels of paranoia (GPTS total; t 120 = 0.642, p = 0.522), severity of the persecutory delusions (PSYRATS total; t 121 = 0.018, p = 0.986), the presence of hallucinations (PANSS hallucinations item; t 121 = −0.263, p = 0.793) or in overall psychopathology (PANSS total; t 121 = −1.028, p = 0.306). However, the patients who showed JTC did have higher levels of negative symptoms (PANSS negative; t 121 = −2.227, p = 0.028).
Neuropsychological functioning and JTC
Those patients who jumped to conclusions, compared with those who did not, showed poorer working memory performance and had lower IQ scores (see Table 2). IQ scores correlated with forward digit span (r = 0.36, p < 0.001), backward digit span (r = 0.50, p < 0.001) and letter–number sequencing (r = 0.55, p < 0.001). Letter–numbering sequencing showed the strongest association with JTC. In a binary logistic regression predicting JTC, when letter–number sequencing and IQ were entered together as independent variables they both became non-significant predictors (p > 0.1). The presence of negative symptoms negatively correlated with forward digit span (r = −0.14, p = 0.129), backward digit span (r = −0.24, p = 0.008), letter–number sequencing (r = −0.24, p = 0.009) and IQ scores (r = −0.37, p < 0.001). In a binary logistic regression predicting JTC, letter–number sequencing remained a significant predictor (p = 0.020), but negative symptoms did not (p = 0.095).
Table 2. Neuropsychological functioning and JTC
JTC, Jumping to conclusions; s.d., standard deviation; CI, confidence interval; IQ, intelligence quotient.
Intolerance of uncertainty and JTC
The total delusion group's mean IUS score of 88.4 (s.d. = 24.1) was high. For example, Buhr & Dugas (Reference Buhr and Dugas2002) reported a mean score in a non-clinical population (n = 273) of 54.8 (s.d. = 17.4), and Norton (Reference Norton2005) reported a mean score in a non-clinical population (n = 540) of 54.9 (s.d. = 18.7). Dudley et al. (Reference Dudley, Shaftoe, Cavanagh, Spencer, Ormrod, Turkington and Freeston2011) reported a mean score of 65.2 (s.d. = 24.6) in patients seen in a first-episode psychosis service (n = 77). The scores of the patients with persecutory delusions are closer to those of a generalized anxiety disorder group. For example, van der Heiden et al. (Reference van der Heiden, Muris and van de Molen2012) reported a mean IUS score of 82.3 (s.d. = 21.1) in a generalized anxiety disorder group (n = 42). There was no difference in the current study in IUS scores between patients with and without hallucinations (t 119 = −1.037, p = 0.302).
Those patients who jumped to conclusions, compared with those who did not, showed lower levels of intolerance of uncertainty, although this was only statistically significant for the presence of inflexible certainty beliefs (see Table 3). Patients with JTC also reported significantly lower levels of worry, but no differences in anxiety or repetitive negative thinking. Worry positively correlated with total IUS scores (r = 0.49, p < 0.001). When total IUS score was entered with worry as independent variables in a binary logistic regression, there was a trend for worry scores to predict JTC (B = −0.055, odds ratio (OR) 0.95, p = 0.067), but IUS scores did not predict JTC (B = −0.008, OR 0.99, p = 0.450). A similar pattern was found for the IUS inflexible certainty beliefs score. In a binary logistic regression predicting JTC, there were indications that worry was a predictor (B = −0.051, OR 0.95, p = 0.078), but the IUS certainty beliefs did not predict JTC (B = −0.073, OR 0.93, p = 0.213).
Table 3. Intolerance of uncertainty and JTC
JTC, Jumping to conclusions; s.d., standard deviation; CI, confidence interval; IUS, Intolerance of Uncertainty Scale.
Working memory, worry and JTC
There were no significant correlations between worry and the working memory tasks (all p > 0.1). In a binary logistic regression with JTC as the dependent variable, and letter–number sequencing and worry as the independent variables, both poorer working memory (B = −0.229, OR 0.80, p = 0.002) and lower levels of worry (B = −0.082, OR 0.922, p = 0.005) predicted JTC. Similarly, in a binary logistic regression, JTC was predicted by both lower IQ (B = −0.037, OR 0.963, p = 0.004) and lower levels of worry (B = −0.076, OR 0.927, p = 0.007). When letter–number sequencing, IQ and worry were simultaneously entered as independent variables, JTC was predicted by worry (B = −0.083, OR 0.921, p = 0.005), but not significantly by working memory (B = −0.160, OR 0.853, p = 0.070) or IQ (B = −0.022, OR 0.978, p = 0.149).
Discussion
Two potential contributors to the JTC reasoning style were examined within a large group of patients with current persecutory delusions. JTC was considered in relation to basic information-processing capacity problems and motivational beliefs concerning uncertainty. It is the former contributor for which the most confident conclusions can be drawn. Given the similar findings of the large studies of Garety et al. (Reference Garety, Joyce, Jolley, Emsley, Waller, Kuipers, Bebbington, Fowler, Dunn and Freeman2013) and Falcone (Reference Falcone2013), we can now be confident of an association of JTC with working memory. Arguably, studies such as that of Ormrod et al. (Reference Ormrod, Shaftoe, Cavanagh, Freeston, Turkington, Price and Dudley2012) failed to find such an association because of a small sample size. The working memory impairment was associated with levels of current intellectual functioning (and these were not separable predictors of JTC). Keeping alternative explanations in mind, gathering data, and weighing the evidence all have plausible memory load; pursuing one explanation, gathering confirmatory evidence, and reaching a definitive conclusion are likely to reduce demands on memory. Working memory difficulties via the reasoning strategies that they provoke are likely to be a factor in the persistence of delusional beliefs.
Levels of intolerance of uncertainty in the patients with persecutory delusions were high, consistent with the presence of worry in this group at a level comparable with patients with anxiety disorders. A number of studies have now identified high levels of worry in patients with persecutory delusions (e.g. Freeman & Garety, Reference Freeman and Garety1999; Startup et al. Reference Startup, Freeman and Garety2007; Bassett et al. Reference Bassett, Sperlinger and Freeman2009; Freeman et al. Reference Freeman, Pugh, Vorontsova, Antley and Slater2010). Against prediction, however, IUS scores were slightly higher in the patients with delusions who did not jump to conclusions compared with the patients who did jump to conclusions. Greater intolerance of uncertainty was clearly not associated with JTC, consistent with the study of Dudley et al. (Reference Dudley, Shaftoe, Cavanagh, Spencer, Ormrod, Turkington and Freeston2011). As originally proposed by Freeston et al. (Reference Freeston, Rhéaume, Letarte, Dugas and Ladouceur1994), intolerance of uncertainty may, at least within a clinical psychosis sample, lead to greater information search before certainty. However, the levels of worry in the group may well explain the association. Caution is needed about over-interpreting this result before replication. This is especially the case since: there was no association with the occurrence of general repetitive negative thinking; an experimental study with patients with persecutory delusions showed no change in JTC in the period after a bout of worry (Freeman et al. Reference Freeman, Startup, Dunn, Černis, Wingham, Pugh, Cordwell and Kingdon2013); and there is little evidence that anxiety disorder groups seek more data during the beads task relative to non-clinical controls (e.g. Jacobsen et al. Reference Jacobsen, Freeman and Salkovskis2012). The study provides little evidence that one form of motivational belief – intolerance of uncertainty – is a factor in JTC.
The key weakness of the study design is that it only identifies factors associated with JTC. An unmeasured variable may better explain the occurrence of JTC. Testing of causal roles is needed, for example, by manipulating working memory performance during data gathering. The group had persistent persecutory delusions, and therefore testing of different delusion types and stages of illness could result in different results. Relationships with other symptom types remain to be established; for example, in the current study there was an association of JTC with negative symptoms, but this was not found in the study of Garety et al. (Reference Garety, Joyce, Jolley, Emsley, Waller, Kuipers, Bebbington, Fowler, Dunn and Freeman2013). There are also other factors that could moderate the extent of data gathering, for example, a reliance on experiential reasoning (Freeman et al. Reference Freeman, Evans and Lister2012 b). As the work is taken forward there will ideally be investigation of data gathering directly in relation to delusional ideas, and hence study of particular reasoning styles that can enhance recovery from persecutory delusions. A greater understanding of the reasoning processes that lead to erroneous beliefs becoming corrected is a clear priority for the development of a new generation of interventions for delusions.
Acknowledgements
The study was supported by a grant (no. 09/160/06) from the Efficacy and Mechanism Evaluation (EME) Programme, which is funded by the UK Medical Research Council and the UK NHS National Institute for Health Research. D.F. is supported by a UK Medical Research Council Senior Clinical Fellowship.
Declaration of Interest
None.
