An implicit measure of associations with mental illness versus physical illness: response latency decomposition and stimuli differential functioning in relation to IAT order of associative conditions and accuracy.

Stefania Mannarini,Marilisa Boffo

doi:10.1371/journal.pone.0101911

Abstract

The present study aimed at the definition of a latent measurement dimension underlying an implicit measure of automatic associations between the concept of mental illness and the psychosocial and biogenetic causal explanatory attributes. To this end, an Implicit Association Test (IAT) assessing the association between the Mental Illness and Physical Illness target categories to the Psychological and Biologic attribute categories, representative of the causal explanation domains, was developed. The IAT presented 22 stimuli (words and pictures) to be categorized into the four categories. After 360 university students completed the IAT, a Many-Facet Rasch Measurement (MFRM) modelling approach was applied. The model specified a person latency parameter and a stimulus latency parameter. Two additional parameters were introduced to denote the order of presentation of the task associative conditions and the general response accuracy. Beyond the overall definition of the latent measurement dimension, the MFRM was also applied to disentangle the effect of the task block order and the general response accuracy on the stimuli response latency. Further, the MFRM allowed detecting any differential functioning of each stimulus in relation to both block ordering and accuracy. The results evidenced: a) the existence of a latency measurement dimension underlying the Mental Illness versus Physical Illness - Implicit Association Test; b) significant effects of block order and accuracy on the overall latency; c) a differential functioning of specific stimuli. The results of the present study can contribute to a better understanding of the functioning of an implicit measure of semantic associations with mental illness and give a first blueprint for the examination of relevant issues in the development of an IAT.

Highlights

Research on the aetiology of psychiatric disorders has vastly expanded our knowledge on the genetic and neurobiological underpinning of mental illness [1]; on the other hand, the research on the aetiology of specific disorders, such as depression and more in general mood disorders, has pointed out the relevance of affective and social factors [2]
The 22 stimuli were considered as indicators of the latent dimension underlying the IAT, representing the speed of association of the Mental Illness and Physical Illness categories with the Psychological and Biologic domains (Mental Illness versus Physical Illness-Implicit Association Test - MIPI-IAT)
Accuracy to the Stimuli Response Latency Considering the response accuracy, the results showed the effect on the speed of categorization of generally performing the task more or less accurately, with a low accuracy associated to faster reaction times

Summary

Introduction

Research on the aetiology of psychiatric disorders has vastly expanded our knowledge on the genetic and neurobiological underpinning of mental illness [1]; on the other hand, the research on the aetiology of specific disorders, such as depression and more in general mood disorders, has pointed out the relevance of affective and social factors [2]. The origin of mental illness, i.e., the causes to which the condition is causally attributed, has been hypothesized to be one of the main components underlying stigmatizing processes towards mental illness [7,8] and has been used as a promotional medium to overcome stigma in a number of public health programs aimed at combating discrimination (e.g., [9,10]). The promotion of biogenetic aetiological beliefs about mental illness has been deemed as a promising approach to reduce stigma, for this type of beliefs is connected to the perception of onset and offset controllability for the stigmatized condition This perspective is well explained within the framework of attribution theory, which holds that the causal attribution of one’s behaviours leads to characteristic emotional, attitudinal, and behavioural responses towards the person in question [14]. Notwithstanding, campaigns sponsoring a biogenetic origin of mental illness did not fully produce the intended effects, with a mixed pattern of negative and positive results (e.g., higher levels of negative stigma but higher endorsement of professional mental health treatments and mental health literacy) [15,16,17,18,19,20,21]; whereas the consequences of the endorsement of a psychosocial aetiology of mental illness remained unchanged in the last 20 years [13]

Methods

Results

Discussion

Conclusion