Portal de Periódicos da CAPES

Delusions and prediction error: re-examining the behavioural evidence for disrupted error signalling in delusion formation

2014; Taylor & Francis; Volume: 19; Issue: 5 Linguagem: Inglês

10.1080/13546805.2014.897601

1464-0619

Oren Griffiths, Robyn Langdon, Mike E. Le Pelley, Max Coltheart,

Psychosomatic Disorders and Their Treatments

AbstractIntroduction. There is now significant evidence that prediction error signalling is mediated by dopamine in the midbrain, and that dopamine dysfunction is implicated in people experiencing psychotic symptoms, including delusions. There has also been significant theorizing and experimentation concerning the remaining link in this triad, namely that deviant prediction error signalling produces or maintains psychotic symptoms.Methods. The research supporting the link between prediction error signalling and delusional symptoms was reviewed. Numerous studies indirectly support this link, but only one set of studies claim to directly test this hypothesis by combining three crucial elements: a patient sample, a manipulation of prediction error and neuroimaging. This particular set of studies were examined in detail.Results. Important methodological limitations in these studies were observed, and a reinterpretation of their data was offered.Conclusions. Methodological inconsistencies significantly weaken the claims made by these studies, but their data are consistent with current theorizing and they are instructive for future lines of inquiry in this field.Keywords: delusionsschizophreniaprediction errorassociative learningpsychopathologyfMRI FundingThis work was supported by an Australian Research Council Discovery Project [grant number DP140104394].Notes1. Although we will use the term "prediction" throughout, many of these predictions are likely preconscious.2. The prodrome is the period that precedes illness onset and is characterised by a marked deviation from a person's normal level of functioning (Yung & McGorry, Citation1996).3. "Pre-pulse inhibition" refers to an attenuation of the normal startle response to an intense stimulus (e.g. a loud noise) due to immediately preceding presentation of a weaker stimulus.4. "Negative" symptoms are termed thus and contrasted with positive psychotic symptoms because of the absence of something that ought normally to be present, such as occurs with apathy.5. Although we will not discuss this issue further, it must be noted that the statistics used to draw these conclusions were not appropriately corrected for multiple comparisons. Failure to do so has been shown to produce dramatically misleading inferences (e.g. Bennett, Baird, Miller, & Wolford, Citation2005).6. Note that Corlett et al. declined to report behavioural data for the high-dose ketamine group (which demonstrated delusion-like symptoms), and instead only reported test data for the low-dose group, which did not display these symptoms.7. Strictly, the prediction that a larger error term on C– trials than on A+ trials relies on the presence of a common context in the Rescorla–Wagner model (it needs this assumption for many basic associative phenomena, e.g. contingency effects, see Rescorla & Wagner, Citation1972). There is also significant empirical evidence from similar studies, including that upon which the design is based (Larkin et al., Citation1998) that the transition from AB+ to A+ training results in a smaller change in outcome predictions than does the CD+ to C– transition (see also Lovibond, Been, Mitchell, Bouton, & Frohardt, Citation2003).8. This contrast was presented by Corlett et al. (Citation2004) as an interaction, comparing the influence of violation vs. confirmation for the unovershadowing condition with that for the backwards blocking condition. That is, the contrast calculates (xDminus − xDplus) − (xBplus − xBminus), where xDminus represents the BOLD response to D– trials in the test phase, xDplus represents the response to D+ trials, etc. Note, however, that this contrast can also be expressed as (xDminus + xBminus) − (xDplus + xBplus), i.e. a main effect comparison of the BOLD response on non-reinforced trials with that on reinforced trials, which is why it is useful for the analysis of our alternative interpretation of these data.Additional informationFundingFunding: This work was supported by an Australian Research Council Discovery Project [grant number DP140104394].