Full Length Article
Jumping to conclusions in the less-delusion-prone? Further evidence from a more reliable beads task

https://doi.org/10.1016/j.concog.2020.102956Get rights and content

Highlights

  • Is delusion-proneness associated with the jumping to conclusions (JTC) bias?

  • A reliable version of the ‘beads task’ was used to clarify this relationship.

  • People scoring high on delusion-proneness requested more evidence (less JTC).

  • Increased odds-literacy did not drive delusion-proneness or JTC.

  • The JTC bias may not contribute to non-clinical delusion-proneness.

Abstract

Background

A single meta-analysis has found that healthy people with higher delusion-proneness tend to gather less information (i.e., make fewer draws to decision, or DTD) on the beads task, although the findings of contributing studies were mixed, and the pooled effect size was small. However, using a new and more reliable “distractor sequences” beads task, we recently found a positive relationship between delusion-proneness and DTD in a healthy sample. In the current study, we re-tested this relationship in a new sample, and tested the possibility that the relationship is driven by participant’s ability to understand and use odds or likelihood information (“odds literacy”).

Methods

Healthy participants (N = 167) completed the distractor sequences beads task, the Peters Delusions Inventory (PDI) which measures delusion-proneness, a measure of odds literacy, and the Depression, Anxiety, and Stress scale.

Results

PDI and DTD were positively correlated, and comparing PDI quartiles on DTD confirmed a statistically significant trend of increasing DTD with PDI quartile. Odds literacy was positively rather than negatively associated with both DTD and PDI. Anxiety was positively correlated with PDI and DTD.

Conclusions

We replicated our earlier finding that DTD and delusion-proneness were positively related in a non-clinical sample, but found that increased odds-literacy did not drive lower PDI and DTD, and hence did not explain their covariance. It is possible however that anxiety and co-occurring risk aversion drive increased delusion-proneness and information-gathering, potentially accounting for the positive relationship between PDI and DTD.

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)

  • R. Raghunathan et al.

    All negative moods are not equal: Motivational influences of anxiety and sadness on decision making

    Organizational Behavior and Human Decision Processes

    (1999)
  • V.F. Reyna et al.

    The importance of mathematics in health and human judgment: Numeracy, risk communication, and medical decision making

    Learning and Individual Differences

    (2007)
  • S.H.-W. So et al.

    ‘Jumping to conclusions’ data-gathering bias in psychosis and other psychiatric disorders—Two meta-analyses of comparisons between patients and healthy individuals

    Clinical Psychology Review

    (2016)
  • J. Van Os et al.

    Strauss (1969) revisited: A psychosis continuum in the general population?

    Schizophrenia Research

    (2000)
  • T.S. Woodward et al.

    Change in delusions is associated with change in “jumping to conclusions”

    Psychiatry Research

    (2009)
  • J. Crawford et al.

    Percentile norms and accompanying interval estimates from an Australian general adult population sample for self-report mood scales (BAI, BDI, CRSD, CES-D, DASS, DASS-21, STAI-X, STAI-Y, SRDS, and SRAS)

    Australian Psychologist

    (2011)
  • R. Dudley et al.

    ‘Jumping to conclusions’ in first-episode psychosis

    Early intervention in psychiatry

    (2011)
  • R. Dudley et al.

    Psychosis, delusions and the “Jumping to Conclusions” reasoning bias: A systematic review and meta-analysis

    Schizophrenia Bulletin

    (2016)
  • J.S.B. Evans et al.

    Dual-process theories of higher cognition: Advancing the debate

    Perspectives on Psychological Science

    (2013)
  • Cited by (0)

    View full text