Full Length ArticleJumping to conclusions in the less-delusion-prone? Further evidence from a more reliable beads task
Introduction
A number of cognitive biases appear related to, and possibly contribute to, delusions in people with psychosis. Of these, evidence is strongest for the jumping to conclusions (JTC) bias, in which people with active delusions seek less information before making a decision than people without active delusions (Dudley et al., 2016, McLean et al., 2017, So et al., 2016).
The JTC bias is typically measured by the beads task (Huq, Garety, & Hemsley, 1988) or its variants such as the ‘fish in lakes’ task (Woodward, Munz, LeClerc, & Lecomte, 2009). In this task participants are presented with two jars, each containing beads of two colours (e.g., colour ‘A’ and colour ‘B’), present in equal but opposite ratios (e.g., 85:15). Participants are informed that one jar has been chosen at random, and beads are drawn one at a time from the chosen jar in a supposedly random, but in fact pre-determined sequence (i.e., AAABAAAABA). As each new bead is drawn, participants may decide which jar beads were being drawn from, or request to see another bead. The number of beads drawn before a decision is made, or “draws to decision” (DTD), operationalises participants’ tendency to gather information before making a decision. In addition, JTC is usually defined as a DTD of only one or two beads.
Delusions and delusion-like ideas are theorised to occur on a continuum (Freeman, 2006, Van Os et al., 2000), and in both clinical and non-clinical populations. Hence it is reasonable to assume that, given JTC is associated with delusions in clinical populations, it should also be associated with delusion-proneness in non-clinical populations. A meta-analysis supported this expectation of increased JTC with higher delusion-proneness (Ross, McKay, Coltheart, & Langdon, 2015). However, the pooled effect size was small, with considerable variation between the individual studies (e.g., over 20% of the included studies reported a trend where JTC was associated with lower delusion-proneness). Moreover, one of the largest general-population samples to date did not find a significant PDI-DTD relationship (Ross et al., 2016).
Our group has recently developed a novel ‘distractor’ version of the beads task (see McLean, Mattiske, & Balzan, 2018), where the single pseudo-random “target” sequence used for measuring DTD (i.e., AAABAAAABA) is surrounded by three “distractor” sequences (i.e., AABAAABAAA, BAAAABAAAA, and AAAABAABAA). These modifications help to ensure that participants do not recognise the repetitious nature of the target sequence when the beads task is administered repeatedly for repeated-measures studies, or for calculating an aggregate measure of DTD. However, in our recent study using this more reliable version of the beads task in a large non-clinical community sample (N = 203), JTC was associated with lower delusion-proneness (McLean, Mattiske, & Balzan, 2020). Specifically, we found that delusion-proneness (quantified using the 21-item Peters et al. Delusions Inventory, or PDI; Peters, Joseph, Day, & Garety, 2004) was positively correlated with DTD, rS = 0.16. We also found that those who jumped to conclusions were lower on delusion-proneness than those who did not JTC. The direction of this result was unexpected, as it contradicts the findings of the Ross et al. (2015) meta-analysis, as well as the prediction of a negative PDI-DTD relationship made by extrapolating the findings of clinical research. Hence the first aim of this study was to test the PDI-DTD relationship in a second non-clinical sample.
Our second aim was to test a potential mechanism for the positive relationship, if replicated, between PDI and DTD. One such mechanism, proposed by McLean et al. (2020), was the ability to understand and apply odds or likelihood information to decision-making, an ability which we term odds literacy. Similar to health numeracy (Reyna and Brainerd, 2007, Reyna et al., 2009), we envisage odds literacy to cover numerical aspects such as fractions, percentages, frequencies, probabilities, and decimals, applied in odds or likelihood situations. Odds literacy may limit the ability of participants to make appropriate beads-task decisions in-so-far as the beads task is probabilistic in nature. Higher odds literacy may drive both fewer DTD, and reduced delusion-proneness, as follows.
Most studies report that non-clinical samples are unnecessarily conservative on the beads task (i.e., request more beads than is required). For example (and quite typically), Jacobsen, Freeman, and Salkovskis (2012) reported healthy-participant DTD of M = 3.88, SD = 1.86. Much higher DTD have also been reported (see for example Menon, Addington, & Remington, 2013, M = 7.88, SD = 2.8). Yet drawing fewer beads is justifiable in probabilistic terms – seeing the evidence of two same-colour beads drawn from jars with an 85:15 ratio allows for choosing a jar with 97% confidence (Fear & Healy, 1997), with confidence rising to 99% by the third bead. It seems possible that more odds-literate people, with a better understanding of the degree of certainty that two or three same-colour beads provides, would choose fewer beads than the average, while less odds-literate people may choose more. In this sense, it could be argued it is participants who chose more beads, and who do not JTC, who are biased (Maher, 1992).
Hence, those with better odds-literacy may choose fewer beads. Concurrently, as delusional thinking involves the acceptance of ideas that are unlikely to be true, more odds-literate individuals may be less likely to entertain delusion-like ideas. If odds literacy drives fewer DTD and concurrently drives lower delusion-proneness, it may partially or fully explain the positive relationship observed between PDI and DTD.
In this study, we re-tested the relationship between delusion-proneness and data-gathering in a non-clinical sample, again using the reliable distractor-sequences beads task. We also tested the possibility that the positive relationship between DTD and delusion-proneness was driven by odds literacy.
Section snippets
Participants
Participants were recruited via a university experiment participation portal. The experiment was open to participants seeking course credit and to participants completing experiments for remuneration, who were paid $15 AUD for participating.
Measures
Draws-to-decision. Our distractor-sequences beads task (McLean et al., 2018) was used to measure DTD. At the beginning of each sequence, target or distractor, participants were shown images of two jars, with each jar containing coloured beads in
Data validation and descriptive statistics
169 participants completed the experiment online. Two participants were excluded for failing the comprehension check more than five times, and final N = 167. Of these, 107 participants completed the experiment for course credit, and 60 for payment. Participants were 141 females and 26 males, aged between 17 and 50 (M = 22.3, SD = 6.8).
Odds literacy (see Table 1) was normally distributed, and appeared to provide good discrimination of participants’ abilities over the possible range of the scale
Replication of the positive PDI-DTD relationship
Our current data replicated our prior finding of a positive correlation between PDI and DTD, and our comparison of participants who jumped to conclusions with those who did not revealed a trend in PDI values in the same direction as our statistically-significant prior finding (McLean et al., 2020). Our data therefore replicate our prior finding that DTD and PDI are positively related in non-clinical samples. This relationship is in the opposite direction to the relationship typically observed
Conclusion
Our replication of McLean et al. (2020) lends confidence to the finding that, when the distractor-sequences beads task is used, PDI and DTD are positively related in healthy samples. This relationship is in the opposite direction to the relationship reported for clinical samples, in which delusions are associated with reduced DTD, that is, with JTC. The positive PDI-DTD relationship in healthy people is not however a consequence of an increased understanding of odds and likelihoods driving
CRediT authorship contribution statement
Benjamin F. McLean: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing - original draft. Ryan P. Balzan: Conceptualization, Formal analysis, Supervision, Writing - review & editing. Julie K. Mattiske: Conceptualization, Formal analysis, Methodology, Supervision, Writing - review & editing.
Acknowledgements
The Authors would like to acknowledge the support of Flinders University, South Australia.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (38)
- et al.
Individualized metacognitive therapy for delusions: A randomized controlled rater-blind study
Journal of Behavior Therapy and Experimental Psychiatry
(2017) - et al.
The psychological effect of an urban environment on individuals with persecutory delusions: The Camberwell walk study
Schizophrenia research
(2008) - et al.
An investigation into the jumping-to-conclusions bias in social anxiety
Consciousness and Cognition
(2017) - et al.
The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories
Behaviour Research and Therapy
(1995) - et al.
Dispositional anxiety and risk-avoidant decision-making
Personality and Individual Differences
(2007) - et al.
The role of risk avoidance in anxiety
Behavior Therapy
(2006) - et al.
Jumping to conclusions in the less-delusion-prone? Preliminary evidence from a more reliable beads task
Journal of Behavior Therapy and Experimental Psychiatry
(2020) - et al.
Towards a reliable repeated-measures beads task for assessing the jumping to conclusions bias
Psychiatry Research
(2018) - et al.
Examining cognitive biases in patients with delusions of reference
European Psychiatry
(2013) - et al.
A two-stage cognitive theory of the positive symptoms of psychosis. Highlighting the role of lowered decision thresholds
Journal of Behavior Therapy and Experimental Psychiatry
(2017)