The role of US recency in the Perruchet effect in eyeblink conditioning

• In the Perruchet design a CS is paired with an airpuff US on 50% of trials.
• Eyeblink CRs and US expectancy show opposite trends over runs of repeated trials.
• We added unpaired USs on CS-alone trials to control for US recency.
• The CR trend disappeared in the control conditions.
• The Perruchet effect does not require postulation of multiple learning systems.
• (All ≤85 characters).

In the Perruchet effect, there is a concurrent dissociation between participants’ conditioned responses (CRs) and their expectancy of the unconditioned stimulus (US) across runs of repeated trials. The effect has been taken as evidence for multiple learning processes, but this conclusion follows only if the CR trend is the result of learning. Two experiments examined the role of US recency in generating the observed CR trend. A standard Perruchet condition was compared with a control condition in which US recency was controlled by presenting the US on every trial. The associative contribution was maintained by varying the temporal relationship between the CS and the US. In both experiments the pattern of CRs seen in the Perruchet condition was absent in the control condition, suggesting that the eyeblink trend in the Perruchet effect may be due to a non-associative performance factor such as priming or sensitization arising from recent US presentations.