Figure 1
Schematic illustration of the two different cognitive mechanisms (task-level inhibition and episodic interference) that could be contributing to N-2 task repetition costs.
Figure 2
Illustration of the N-X contrast as a method for assessing contributions of episodic interference to N-2 task repetition costs. Episodic interference in the current trial N is defined with respect to the last occurrence of the same task in trial N-2 (in ABA sequences) or in trial N-3, N-4, N-5, etc., summarized as N-X (in CBA sequences). The N-X contrast is shown for three different levels of episodic interference between the current task in Trial N and the most recent episode of that task in Trial N-X: Episodic match of task-relevant and task-irrelevant features (upper row); Episodic match of task-relevant features only (i.e., of stimulus category and associated response; middle row); Episodic mismatch of task-relevant features (stimulus category and associated response) and task-irrelevant features (stimulus identity; lower row). Participants switch between three different face categorization tasks: Indicating whether the presented face is female or male (task cue: blue frame around the picture); indicating whether the face belongs to an old or young person (task cue: red frame); indicating whether the face shows an angry or happy expression (task cue: yellow frame). The stimulus set included 40 different pictures overall (each showing a different person; five different persons for each combination of task-relevant categories, e.g., female-young-happy). Participants responded by pressing a left or right response key; the same two keys were used for all three tasks, and the response mappings were counterbalanced across participants (see text for details).
Figure 3
Re-analysis of the young adults’ groups from Schuch & Konrad (2017) and Schuch (2016). N = 56. The condition of Episodic Match of task-relevant and task-irrelevant features was not manipulated in these data. Left side: Mean RT (upper row) and mean Error Rate (lower row) as a function of N-2 Task Transition (N-2 Task Repetition [ABA] versus N-2 Task Switch [CBA]) and Episodic Match Condition (Episodic Match of task-relevant features [stimulus category and response] between current and last task episode; Episodic Mismatch: Switch of the task-relevant features [stimulus category and response] between current and last task episode). Right side: N-2 task-repetition costs as a function of Episodic Match Condition in RTs and Error Rates. Error bars represent the 95% confidence interval of the ABA-CBA difference per Episodic Match Condition (Pfister & Janczyk, 2013).
Figure 4
Experiment 1. N = 40. Left side: Mean RT (upper row) and mean Error Rate (lower row) as a function of N-2 Task Transition (N-2 Task Repetition [ABA] versus N-2 Task Switch [CBA]) and Episodic Match Condition (Full Episodic Match: Repetition of all task features [stimulus identity, stimulus category, and response] between current and last task episode; Episodic Match of task-relevant features [stimulus category and response]; Episodic Mismatch: Switch of the task-relevant features [stimulus category and response]). Right side: N-2 task-repetition costs as a function of Episodic Match Condition in RTs and Error Rates. Error bars represent the 95% confidence interval of the ABA-CBA difference per Episodic Match Condition (Pfister & Janczyk, 2013).
Figure 5
Mean RT (upper row) and mean Error Rate (lower row) as a function of Task Lag and Episodic Match Condition. In Experiment 1, five different levels of Task Lag were distinguished (left side); in the re-analysis of the published data, three different levels of Task Lag were distinguished due to lower trial numbers (right side). Error bars are not shown for better visualization of the data pattern; see Table S3 in Online Supplemental Material for the standard error of mean per condition.