Sequential effects within a short foreperiod context Evidence for the conditioning account of temporal preparation

• Michael B. Steinborn , Bettina Rolke, Daniel Bratzke, Rolf Ulrich

Abstract

• Responses to an imperative stimulus (IS) are especially fast when they are preceded by a warning signal (WS).
• When the interval between WS and IS (the foreperiod, FP) is variable, reaction time (RT) is not only influenced by the current FP but also by the FP of the preceding trial
• sequential effects originate from a trace conditioning process, in which the individuals learn the temporal WS–IS relationship in a trial-by-trial manner
• that trace conditioning is maximal when the temporal interval between the conditioned and unconditioned stimulus is between 0.25 and 0.60 s
• , one would predict that sequential effects occur especially within short FP contexts.
• However, this prediction is contradicted by Karlin [Karlin, L. (1959)]
• investigate temporal preparation for short FPs
• The results provide strong evidence for sequential effects within a short FP context and thus support the trace conditioning account of temporal preparation.

Experiment 1

• Anticipatory responding was controlled by using a choice RT task
• There was a main effect of the factor FP-set on RT, F(1,21) = 219.3, partial g2 = .91, p < .001, indicating that RT was shorter in the short FP-set (366 ms) than in the long FP-set
• RT benefit for the short FP-set might be attributable to a better general ability to process short time intervals than long ones (e.g., Klemmer, 1957; Näätänen et al., 1974).
• main effect of FPn on RT,
• RT decreased as $F{P}_n$ increased
• FP in the preceding trial also influenced RT in the current trial as revealed by a main effect of $F{P}_{n-1}$ on RT
• increased as $F{P}_{n-1}$ decreased
• $F{P}_{n-1}$ $F{P}_n$ interaction effect on RT
• when the preceding FP was long, RT in a current trial decreased with increasing FP and this effect was
• weaker when a short FP preceded a current trial
• asymmetry of the sequential FP effect was smaller for the short FP-set than for the long FP-set.
• Most important, however, the sequential FP effect was not restricted to the long FP-set but was also present for the short FP-set of FP-durations below 0.6 s

Experiment 2

• assessed sequential FP effects in a simple RT task employing only the short FP-set
• catch trial technique
• compared the asymmetrical sequential FP effect in a condition with 0% catch trials (referred to as no-CT condition) to a condition with 25% catch trials (referred to as CT condition)
• RT was prolonged in the CT condition
• main effect of FPn on RT
• decrease of RT with increasing FPn
• $F{P}_{n-1}$ influenced RT
• influence of the preceding FP was unaffected by CT
• asymmetrical sequential FP effect again showed up in the $F{P}_{n-1}$ $F{P}_n$ interaction on RT
• anticipatory responses also increased with decreasing preceding FP
• There was a significant $F{P}_{n-1}$ $F{P}_n$ interaction on RT for the no-CT condition
• weak CT $F{P}_n$ interaction effect on RT

Experiment 3

• FP-range employed in Karlin’s study was too dense and therefore did not produce sufficient temporal uncertainty
• simple RTs were much faster (220 ms) than choice RTs
• RT pattern differed between the task conditions,
• In contrast to Experiment 2, an upwardsloping FP-RT effect was observed
• RT increased from the shortest towards the longest $F{P}_n$
• RT decreased with increasing $F{P}_{n-1}$
• responses in short $F{P}_n$ trials were always fast irrespective of $F{P}_{n-1}$.
• In contrast, responses in long FPn trials were on average slower and showed a sequential modulation.
• Precisely, in long FPn trials, responses were relatively fast when $F{P}_{n-1}$ was also long compared to when $F{P}_{n-1}$ was short.
• In sum, the simple RT condition revealed especially fast responses and an extraordinary high percentage of anticipatory responses in short FPn trials, even though FPn1 was long
• is consistent with the results of Karlin (1959) and suggests that participants mainly prepared for an early imperative moment without re-preparing in long FPn trials
• the RT pattern in the simple RT condition clearly indicates that Karlin’s (1959) finding was not an anomalous result but a reliable empirical phenomenon that occurs when average FPs are small and the FPrange is very dense
• The overall pattern of results (simple and choice RT condition) is consistent with the view that participants already attained maximal preparation at the short $F{P}_n$ and were not able to re-prepare when the IS did not occur at the short $F{P}_n$.
• Instead, they may have relied on residual preparatory activity from the early imperative moment (Alegria, 1974; Alegria, 1975b).
• instance, they may have shifted a single moment of peak preparation in a rather analog way, that is, after a short $F{P}_n$ they expected the IS somewhat earlier, after a long FPn1 $F{P}_{n-1}$ somewhat later

General discussion

• when the FP-range is too dense, the typical asymmetrical sequential FP effect does not occur.
• Hence, the RT pattern observed in the simple RT condition demonstrates that Karlin’s (1959) finding of a reversed sequential FP effect is a robust phenomenon that occurs in simple RT tasks when the FP-range is very dense.
• We suggest that when the FP-range is very dense, the individuals may not represent three distinct imperative moments but a single relatively noisy one to which they attain preparation
• The result pattern of the simple RT condition indicates that participants attained preparation at an early imperative moment because responses were especially fast and a high amount of anticipatory responses were observed in short FPn trials
• The observation of a reversed sequential effect, however, shows that the moment of peak preparation was still influenced by the preceding trial.
• In particular, participants may have expected the IS after a short FPn1 somewhat earlier, but after a long
• $F{P}_{n-1}$ somewhat later in time
• Critically, when the small FP-range does not enable a sharpedged representation of three distinct critical moments but only a noisy representation of a single critical moment, then the process that produces the asymmetrical sequential FP effects at short FP
• no sequential FP effect in short $F{P}_n$ trials should be expected in this situation
• Importantly, a rather analog sequential adjustment of a single but early preparatory peak should result in a sequential modulation at later imperative moments, as is exactly observed in the simple RT condition of Experiment 3
• If temporal preparation had increased with $F{P}_n$-length, this should have resulted in more efficient performance

Images

• 
• 
•