Statistical learning under incidental versus intentional conditions

Abstract

Statistical learning (SL) studies have shown that participants are able to extract regularities in input they are exposed to without any instruction to do so. This and other findings, such as the fact that participants are often unable to verbalize their acquired knowledge, suggest that SL can occur implicitly or incidentally. Interestingly, several studies using the related paradigms of artificial grammar learning and serial response time tasks have shown that explicit instructions can aid learning under certain conditions. Within the SL literature, however, very few studies have contrasted incidental and intentional learning conditions. The aim of the present study was to investigate the effect of having prior knowledge of the statistical regularities in the input when undertaking a task of visual sequential SL. Specifically, we compared the degree of SL exhibited by participants who were informed (intentional group) versus those who were uninformed (incidental group) about the presence of embedded triplets within a familiarization stream. Somewhat surprisingly, our results revealed that there were no statistically significant differences (and only a small effect size) in the amount of SL exhibited between the intentional versus the incidental groups. We discuss the ways in which this result can be interpreted and suggest that short presentation times for stimuli in the familiarization stream in our study may have limited the opportunity for explicit learning. This suggestion is in line with recent research revealing a statistically significant difference (and a large effect size) between intentional versus incidental groups using a very similar visual sequential SL task, but with longer presentation times. Finally, we outline a number of directions for future research.

The Necessity of the Hippocampus for Statistical Learning

“…In Experiment 1, patient L. S. J. performed at chance and significantly worse than matched healthy comparison participants. However, although comparison participants in Experiment 1 performed better than chance at the group level, they demonstrated significant variability, with a sizeable minority performing at chance levels (this variability in performance in healthy adult participants is consistent with the statistical learning literature more broadly; see (Erickson, Kaschak, Thiessen, & Berry, 2016; Arciuli, Torkildsen, Stevens, & Simpson, 2014; Misyak & Christiansen, 2012; Saffran, Newport, Aslin, Tunick, & Barrueco, 1997). Therefore, Schapiro et al implemented Experiment 2, which consisted of repeating pairs of stimuli.…”

“…Furthermore, Patients 1846 and 2363 also have significant reductions in hippocampal volume and severe declarative memory impairment. Although significant variability in statistical learning tasks, both within and among participants, is well documented (Erickson et al, 2016; Arciuli et al, 2014; Misyak & Christiansen, 2012; Saffran et al, 1997), it is difficult to discern why the patients in this study performed consistently better than L. S. J.…”

“…Performance of healthy participants in within-group studies is often compared to chance (e.g., Schlichting, Guarino, Schapiro, Turk-Browne, & Preston, 2017; Arciuli et al, 2014; Toro, Sinnett, & Soto-Faraco, 2005; Creel et al, 2004; Saffran, Johnson, Aslin, & Newport, 1999). Comparing performance to chance levels is also a common metric of learning and memory in the amnesia literature (e.g., Dede, Frascino, Wixted, & Squire, 2016; Schapiro et al, 2014; Hannula et al, 2006).…”

Investigating implicit statistical learning mechanisms through contextual cueing

“…However, signs of learning can appear in the MTL memory system after a handful of repetitions, even in the absence of awareness according to familiarity measurement criteria [46,118]. Moreover, incidental learning can be as robust as intentional learning [119] SL might thus precede awareness of regular structures, instead of being concomitant with it.…”

The multi-component nature of statistical learning

“…Although investigations are ongoing, there is substantial evidence that SL can operate largely implicitly even though certain SL tasks can be modified in ways that result in more or less explicit knowledge; for example, via task instructions (e.g. [12][13][14]). This paper focuses on SL as a largely implicit process.…”