The foreperiod effect revisited: conditioning as a basis for nonspecific preparation1
Section snippets
Basic design and phenomena
In a pilot study, we used three clearly distinct FPs of 0.5, 1.0 and 1.5 s in a compatible two-choice reaction task. We refer to the moments immediately following these FPs as critical moments; these are moments the imperative stimulus could be presented across a given block of trials. We refer to the moment the imperative stimulus is presented, on any specific trial, as the imperative moment. We presented the different FPs in both pure and mixed blocks of trials. In pure blocks the same FP was
The nature of the preparatory process
As noted earlier, the basic FP-effects have been invariably attributed to the participant's state of nonspecific preparation at the imperative moment. It should be obvious, though, that this does not explain anything as long as it is not specified how fluctuations in the preparatory state come about. Regarding this issue, the modal view in the literature is based on two assumptions. First, a high preparatory state is quickly attained but hard to maintain over time (e.g., Gottsdanker, 1975,
Outlines of a conditioning model
To account for the basic FP-effects shown in Fig. 1, a conditioning view of nonspecific preparation demands further specification. The critical issue here is not whether conditioning occurs, but when it occurs, that is, how the CR develops in real time, and why it is stronger at some critical moments than at others. In fact, this is a core issue in trace conditioning – a form of classical or operant conditioning which is defined by the occurrence of a blank interstimulus interval between the CS
A formal conditioning model
Machado (1997) developed a formal model of trace conditioning whose learning rules bear strong similarity to those of Los (1996) presented above. In this section, we first give a verbal description of this model, followed by a formal treatment.
Fig. 2shows the structure of the model (cf. Machado, 1997, p. 241).2
Method
Participants. Nine students with normal or corrected-to-normal vision served as paid participants.
Apparatus. The experiment was conducted on a personal computer, equipped with a 486 processor, and connected to a VGA color monitor. The program ERTS controlled the sequence of experimental events, while the EXKEY interface between the computer and a response panel allowed RT to be measured to the nearest millisecond (Beringer, 1992). The response panel had a 23×37 cm, 10° sloping plane, on which
Results
Observations. The solid lines of Fig. 5show mean RT and mean error percentage as a function of block type, FPi and FPi−1. These data were subjected to univariate analyses of variance (ANOVAs). We applied the Greenhouse–Geisser correction in all analyses to correct for possible violations of sphericity of the variance–covariance matrix of RTs (e.g., Stevens, 1992), and we compared the resulting p-values to an α value of 0.05. A first analysis aimed at assessing effects of block type. For this
Discussion
Judged from the values of R2, the outcomes of our simulations provide only moderate support for the model we adjusted from Machado (1997) as a theory of nonspecific preparation. It stands to reason, however, that high values of R2 are not readily obtained when the data feature low variance, due to which the contribution of unsystematic variance (i.e., noise) becomes disproportionately large. As is clear from Fig. 6, the variance of our data was quite low, especially because RTs for long FPs
Concluding remark
In his original article, Machado (1997) called the timing nodes “behavioral states”, which correspond to behaviors animals typically show in between two moments of reinforcement, such as searching for food, pecking around on the floor, and pacing along the side walls of the chamber. In this final section, we feel that it is appropriate to defend our alternative interpretation of these nodes as timing nodes that constitute an internal clock.
We first note that our typical estimate for λ of about
Acknowledgements
We thank Armando Machado and two anonymous reviewers for helpful comments on earlier drafts.
References (43)
Neural aspects of anticipatory behavior
Acta Psychologica
(1999)- et al.
A neural model of timed response learning in the cerebellum
Neural Networks
(1994) - et al.
Further argument for the existence of a pacemaker in the human information processing system
Acta Psychologica
(1997) - et al.
A neural network model of adaptively timed reinforcement learning and hippocampal dynamics
Cognitive Brain Research
(1992) - et al.
Neural dynamics of adaptive timing and temporal discrimination during associative learning
Neural Networks
(1989) Origins of behavior in Pavlovian conditioning
The Psychology of Learning and Motivation
(1984)On the origin of mixing costs: exploring information processing in pure and mixed blocks of trials
Acta Psychologica
(1996)Nonaging foreperiod and simple reaction time
Acta Psychologica
(1971)Foreperiod duration and the time course of preparation
Acta Psychologica
(1972)Issues and trends in the debate on discrete vs. continuous processing of information
Acta Psychologica
(1990)
The discovery of processing stages: extensions of Donders' method
Acta Psychologica
The time course of preparation
Quarterly Journal of Experimental Psychology
Waiting in readiness: gating in attention and motor preparation
Psychophysiology
Effects of facilitory and inhibitory sets on GSR conditioning and performance
Journal of Experimental Psychology
Information processing and human autonomic classical conditioning
Advances in Psychophysiology
Heart rate: changes during conditioned suppression in rats
Science
The biological function of Pavlovian conditioning: the best defence is a good offense
Journal of Experimental Psychology: Animal Behavior Processes
Time uncertainty in simple reaction time
Journal of Experimental Psychology
Cited by (0)
- 1
Parts of this article were presented on the symposium “Looking for stages”, organized on the occasion of Andries F. Sanders becoming Professor Emeritus.