Costly habitual avoidance is reduced by concurrent goal-directed approach in a modified devaluation paradigm

Rethinking avoidance

“…The lack of habitual avoidance in our study in combination with the null findings regarding habitual avoidance by Glück et al (2021) might indicate that habitual avoidance behaviors do not exist. However, given that some studies did demonstrate habitual avoidance responses, it could also imply that habitual avoidance responses can only be observed under special conditions (e.g., high complexity or time pressure).…”

“…The null findings regarding habitual avoidance in experiment 2 by Glück et al (2021) suggest that the complexity of the task affects the expression of habitual avoidance. Comparing experiments 1 and 2 of Glück and colleagues (2021), it was concluded that the strength of avoidance habits was modulated by the complexity and ambiguity of the competing goaldirected approach behavior.…”

“…Moreover, avoidance habits were primarily strong immediately after devaluation (i.e., in the first block) and accuracy was above chance level in the first block reaching high overall response accuracy in the third block, causing doubt on whether we should consider the persistent responding malaptive. In their second experiment, participants were instructed about the changing stimulus-action-outcome contingencies in order to disentangle habitual responding from strategy-driven responding, which resulted in a failure to find evidence in favor of costly habitual avoidance (Glück et al, 2021).…”

“…Otto et al (2013) demonstrated that performing a cognitively demanding task leads to increased reliance on seemingly habitual behavior. Thus, in more complex tasks (e.g., Glück et al, 2021; experiment 1) optimal responding requires more effort and suboptimal behavior (or habitual behavior) is more readily displayed. Our study (Chapter 2) also used a relatively simple habitual avoidance task.…”

“…Moreover, as is the case with fear conditioning research in general, many variations and subtle differences in procedures exist that, often unintentionally, engage different processes, resulting in different outcomes and a difficulty to compare findings between studies . To illustrate, our novel avoidance reversal learning paradigm is different from the outcome devaluation procedure that is typically used when investigating habitual avoidance (e.g., Glück et al, 2021;Gordon et al, 2020;. Future studies are needed that test and validate research designs that are able to induce avoidance habits, if they even exist, ultimately reaching consensus on optimal ways to experimentally study the maladaptive (habitual) avoidance displayed by anxious individuals.…”