Implicit cognition and substance use: the role of controlled and automatic processes in children

Abstract

Objective: Most cognitive models of substance abuse and dependence posit that controlled and automatic processes are central to substance use. Tests of these models rely on methods that are interpreted to measure one or the other of these processes. There has been growing interest in the use of implicit substance use tasks, which are posited to reflect automatic processes. Recent model advancements suggest that behavior is determined by multiple cognitive processes and that dual-process models may provide an overly simplistic account of the cognitive process involved in the assessment of implicit cognition. The goal of the current study was to apply the Quad Model to children's performance on implicit substance use tasks and consider associations with early substance use.

Method: Children (N = 378; 52% girls) ranging from 10 to 12 years old completed alcohol and cigarette Single Category Implicit Association Tests (SC-IATs) and self-reports of substance use.

Results: Four distinct cognitive processes were found to influence SC-IAT performance, one of which reflected automatic activation, the process typically viewed as central to IAT performance. Differences across drinking status revealed weaker automatic activation of negative alcohol associations for those who had (vs. had not) initiated drinking, and a strong likelihood to overcome biased attitudes was supported for all children. The low prevalence of cigarette use in our young sample prohibited examination of the model across smoking status.

Conclusions: Findings suggest that performance on implicit substance use tasks is not process pure. Quantifying and interpreting the multiple influencing processes are crucial for further development and evaluation of cognitive risk models of substance use.

Figure 1

Processing tree of the quadruple process model (Quad Model) applied to substance use (alcohol, cigarettes). Model presented shows likelihood of correct/incorrect responses when there is activation of a negative substance association and general positive category (i.e., negative substance use pairing is compatible). Parameters with lines leading to it are conditioned on all preceding parameters. Right side of figure shows response [correct (+) or incorrect (−)] to valenced (positive and negative) words and substance use (alcohol, cigarettes) targets as a function of processes and block type (compatible; incompatible). AC = activation of an association; D = detection (correct response); OB = overcoming bias; G = guessing right key; 1 -AC = lack of activation of an association; 1-D = failure of detection (correct response); 1-OB = failure to overcome bias; 1-G = guessing left key; Pos. = positive; Neg. = negative.