Does temporal contiguity moderate contingency learning in a speeded performance task?

Abstract

In four experiments, we varied the time between the onset of distracting nonwords and target colour words in a word–word version of the colour–word contingency learning paradigm. Contingencies were created by pairing a distractor nonword more often with one target colour word than with other colour words. A contingency effect corresponds to faster responses to the target colour word on high-contingency trials (i.e., distractor nonword followed by the target colour word with which it appears most often) than on low-contingency trials (i.e., distractor nonword followed by a target colour word with which it appears only occasionally). Roughly equivalent-sized contingency effects were found at stimulus-onset asynchronies (SOAs) of 50, 250, and 450 ms in Experiment 1, and 50, 500, and 1,000 ms in Experiment 2. In Experiment 3, a contingency effect was observed at SOAs of –50, –200, and –350 ms. In Experiment 4, interstimulus interval (ISI) was varied along with SOA, and learning was equivalent for 200-, 700-, and 1,200-ms SOAs. Together, these experiments suggest that the distracting stimulus does not need to be presented in close temporal contiguity with the response to induce learning. Relations to past research on causal judgement and implications for further contingency learning research are discussed.

Keywords:

• Contingency learning
• Temporal contiguity
• Performance tasks
• Interstimulus interval
• Stimulus onset asynchrony

Acknowledgments

James R. Schmidt is a postdoctoral fellow of the Scientific Research Foundation–Flanders (FWO). This research was supported by a Visiting Postdoctoral Fellowship from the Research Foundation of Flanders (FWO) to James R. Schmidt and by Grants BOF/GOA2006/001 and BOF09/01M00209 of Ghent University to Jan De Houwer. We would like to thank Tom Beckers for his helpful comments.

Notes

¹ Other explanations have been forwarded and tested. For instance, one could argue that participants simply respond faster to high- than to low-contingency stimuli because they are more visually familiar. However, Schmidt and colleagues (2007) demonstrated that this cannot be the case, because learning is not dependent on the stimuli that are presented, but instead is dependent on whether the distractor is accurately predictive of the response. Further, Schmidt and colleagues (2010) demonstrated that the effect is not driven by the repetition of stimulus features across trials, but instead to the overall statistical contingency inferred across trials.

² In fact, our initial attempts at studying the effect of contiguity on contingency learning using such a flanker paradigm were generally unsuccessful in producing a contingency effect at any lag.

³ There is one important caveat to highlight with regard to the proximal onset hypothesis. In most contiguity research, SOA is manipulated by presenting fixed-duration stimuli (e.g., a 100-ms tone) and varying the response–stimulus interval (RSI) or interstimulus interval (ISI). We used this approach in Experiment 4, but in the first three experiments we instead used continuously presented stimuli and only manipulated SOA. Thus, the distractor and target always overlapped in presentation time regardless of SOA condition. It could therefore be argued that this may not represent a true manipulation of temporal contiguity. However, we thought it highly plausible when starting this research that variations in stimulus onsets might in fact have an impact on what is learned, especially given that many effects in cognitive psychology are strongly influenced by such manipulations (e.g., the Stroop effect; Glaser & Glaser, 1982). Regardless, the results of Experiment 4 echo those of the first three experiments.

⁴ There were some clear response time outliers that exaggerated this correlation somewhat, but trimming them did not eliminate the correlation entirely.

⁵ It is also possible that the proximal onset and preview advantage hypotheses are both correct, and the two counteract each other, producing an apparent null effect. However, it seems somewhat unlikely that the two would perfectly counteract each other in all four experiments to produce an additive relationship between lag and contingency.

⁶ There is some debate as to whether hidden covariation detection is a genuine (or at least generalizable) effect. See Hendrickx, De Houwer, Baeyens, Eelen, and Van Avermaet (1997a, 1997b) and Lewicki, Hill, and Czyzewska (1997) for more on this debate.