Previous work has found that the contingencies experienced at UK traffic lights can affect drivers behavior potentiallyleading to risky driving. However, these studies did not account for the sequences experienced at traffic lights. Thisexperiment seeks to rectify this. As with previous research we used an incidental go/no-go task in which colored shapeswere stochastically predictive of whether a response was required. The stimuli encoded the contingencies of traffic lightsand their appropriate response, for example, stimuli G was a go cue, mimicking the response to a green light. Crucially,cues were displayed in the sequences experienced at traffic lights. Supporting earlier work, the 50/50 cue that mimickedamber traffic lights was experienced as a go cue, and the stop cue that represented red lights was responded to as a neutralcue. The sequences seemed to enhance this pattern of learning with much larger effect sizes than previously found.

The reaction time (RT) version of the Perruchet Effect is basedon a concurrent dissociation between RTs to respond andconscious expectancy of the outcome across runs of repeatedtrials. Consequently, the Perruchet Effect is considered strongevidence for multiple learning processes. This conclusion,however, relies on the RT trend being driven by associativelearning rather than, as some have argued, US recency orpriming mechanisms. Recent research examining themechanisms underlying the RT trend do so by examiningmotor activity associated with the response. With this aim inmind, the current study developed, and assessed the usefulnessof, a novel method to measure changes in the amount of forceapplied to the response button in an RT Perruchet paradigm.The results obtained could not be explained by a singlemechanism, but suggest multiple factors underlying the RTversion of the Perruchet effect.

To improve road safety, it is important to understand the impact that the contingencies around traffic lights have upon drivers’ behavior. There are formal rules that govern behavior at UK traffic lights (see The Highway Code, 2015), but what does experience of the contingencies do to us? While a green light always cues a go response and a singleton red a stop, the behavior linked to amber is ambiguous; in the presence of red it cues readiness to start, while on its own it cues "preparation" to stop. Could it be that the contingencies between stimuli and responses lead to implicit learning of responses that differ from those suggested by the rules of the road? This study used an incidental go/no-go task in which colored shapes were stochastically predictive of whether a response was required. The stimuli encoded the contingencies between traffic lights and their appropriate responses, for example, stimulus G was a go cue, mimicking the response to a green light. Evidence was found to indicate that G was a go cue, while A (which had the same contingencies as an amber light) was a weak go cue, and that R (a stop cue) was surprisingly responded to as a neutral cue.