Dambacher, M. 1, 2 & Hübner, R. 1
1 Cognitive Psychology, University of Konstanz, Germany
2 Zukunftskolleg, University of Konstanz, Germany
Deadlines (DL) and response signals (RS) are two well-established tools for the investigation of speed-accuracy trade-offs (SATs). While DL require responses before a pre-specified period has elapsed, responses to RS must be made immediately after an explicit signal. Thus, the detection of an additional signal in RS paradigms potentially reduces performance in the main task. Here, we compared the two techniques in a flanker task where participants indicated the parity of target digits in the presence of neutral or response-incompatible flankers. Five response intervals with upper limits from 375 to 750 ms were used in separate DL and RS sessions. Both methods yielded robust SAT functions, i.e., faster and less accurate responses for shorter intervals. Overall, the range of response times was more extended with RS. Yet, response times for late RS showed a bimodal distribution, and the flanker effect, as indicated by faster and more accurate responses for neutral than for incompatible stimuli, was smaller with RS than with DL. The data suggest that responses to RS are not a pure measure of the current state of stimulus processing but partly reflect waiting periods for the signal, potentially limiting RS-based inferences about evidence accumulation.