Stroop interference depends also on the level of automaticity of the to-be-interfered process.

The size of the Stroop effect is usually taken as dependent on the level of practice of the more automatized of two competing processes (e.g., reading in the standard Stroop task), possibly modulated in children by the age-dependent ability to inhibit nonrelevant information. However, this conclusion stems from experimental settings where the automaticity of the second process (e.g., color naming) is hard to assess and manipulate. The musical Stroop task, in which a note name is written inside a note on a staff, overcomes this limit. In the present experiment, children engaged in musical education were asked to read the written note names while ignoring the notes on the staff, or conversely, to name the notes while ignoring the written names. Both a Stroop-like effect and its reverse were observed, but, unexpectedly, the two effects did not evolve in parallel even though both musical and reading abilities improved during practice. Introducing the level of immunity to interference of the to-be-interfered process as a predictor of Stroop interference, in addition to the strength of the interfering process, appears as the best way to account for the interactive pattern.

What Mechanisms Underlie Implicit Statistical Learning? Transitional Probabilities Versus Chunks in Language Learning.

In a prior review, Perrruchet and Pacton (2006) noted that the literature on implicit learning and the more recent studies on statistical learning focused on the same phenomena, namely the domain-general learning mechanisms acting in incidental, unsupervised learning situations. However, they also noted that implicit learning and statistical learning research favored different interpretations, focusing on the selection of chunks and the computation of transitional probabilities aimed at discovering chunk boundaries, respectively. This paper examines the state of the debate 12 years later. The link between contrasting theories and their historical roots has disappeared, but a number of studies were aimed at contrasting the predictions of these two approaches. Overall, these studies strongly question the still prevalent account based on the statistical computation of pairwise associations. Various chunk-based models provide much better predictions in a number of experimental situations. However, these models rely on very different conceptual frameworks, as illustrated by a comparison between Bayesian models of word segmentation, PARSER, and a connectionist model (TRACX).

Familiar units prevail over statistical cues in word segmentation.

In language acquisition research, the prevailing position is that listeners exploit statistical cues, in particular transitional probabilities between syllables, to discover words of a language. However, other cues are also involved in word discovery. Assessing the weight learners give to these different cues leads to a better understanding of the processes underlying speech segmentation. The present study evaluated whether adult learners preferentially used known units or statistical cues for segmenting continuous speech. Before the exposure phase, participants were familiarized with part-words of a three-word artificial language. This design allowed the dissociation of the influence of statistical cues and familiar units, with statistical cues favoring word segmentation and familiar units favoring (nonoptimal) part-word segmentation. In Experiment 1, performance in a two-alternative forced choice (2AFC) task between words and part-words revealed part-word segmentation (even though part-words were less cohesive in terms of transitional probabilities and less frequent than words). By contrast, an unfamiliarized group exhibited word segmentation, as usually observed in standard conditions. Experiment 2 used a syllable-detection task to remove the likely contamination of performance by memory and strategy effects in the 2AFC task. Overall, the results suggest that familiar units overrode statistical cues, ultimately questioning the need for computation mechanisms of transitional probabilities (TPs) in natural language speech segmentation.

Dissociating conscious expectancies from automatic-link formation in an electrodermal conditioning paradigm.

The key point of a paradigm initially proposed by Perruchet (Pavlov J Biol Sci 20:163-170, 1985) to dissociate conscious expectancies from automatic-link formation in classical conditioning settings is the use of a partial reinforcement schedule, in which the unconditioned stimulus (US) follows the conditioned stimulus (CS) only half of the time on average. Given (pseudo) randomization, the whole sequence comprises runs of CS alone and runs of CS-US pairs of various lengths. When the preceding run goes from a long sequence of CS alone to a long sequence of CS-US pairs (via shorter sequences), associative strength should grow up, whereas conscious expectancy should decrease. Earlier studies have shown that, in most cases, conditioned performance parallels associative strength. As an exception, however, a few reports suggest that conditioned electrodermal responses (EDRs) would follow predicted changes in US expectancies. This paper presents an experiment that replicates this outcome. However, when the performances from a control group were taken as a baseline to control for response habituation, corrected conditioned EDRs were shown to follow associative strength. This suggests that the atypical pattern of conditioned EDRs in the Perruchet paradigm would be due to the fact that EDRs are more sensitive to habituation than responses involved in other associative learning settings. These results further challenge the recent "propositional" view of conditioning, which stipulates that conditioned responses in humans are the consequence of participants' conscious inferences about the relationships between the CS and the US, which would lead the CS to generate conscious expectancy for the US.

Dissociating conscious expectancies from automatic link formation in associative learning: a review on the so-called Perruchet effect.

A long-running debate in the literature on conditioning in humans focuses on the question of whether conditioned responses are the product of automatic link formation processes governed by the standard laws of simple associative learning, or the consequence of participants' inferences about the relationships between the 2 related events, E1 and E2, which would lead E1 to generate a conscious expectancy of E2. A paradigm aimed at dissociating the predictions of the 2 accounts was proposed by Perruchet (1985). In this paradigm, E2 randomly follows E1 only half of the time on average, a probability that is known to participants. When the preceding run goes from a long sequence of E1 alone to a long sequence of E1-E2 pairs, associative strength should increase, whereas conscious expectancy for E2 should decrease in keeping with the gambler's fallacy. This article reviews the studies making use of the paradigm in the classical conditioning domain, and the extension of the same logic to a few other experimental situations. Overall, overt behavior has been found to change in line with associative strength, and in opposition to conscious expectancy, attesting to an empirical dissociation of automatic and control processes within a single preparation. The paradigm, however, is endowed with a number of tricky methodological issues, which are examined each in turn. Although some of these issues call for further research, a tentative conclusion is that the effect provides evidence for automatic link formation processes, the existence of which has been recently denied in the "propositional" account of learning.

Is an attention-based associative account of adjacent and nonadjacent dependency learning valid?

Pacton and Perruchet (2008) reported that participants who were asked to process adjacent elements located within a sequence of digits learned adjacent dependencies but did not learn nonadjacent dependencies and conversely, participants who were asked to process nonadjacent digits learned nonadjacent dependencies but did not learn adjacent dependencies. In the present study, we showed that when participants were simply asked to read aloud the same sequences of digits, a task demand that did not require the intentional processing of specific elements as in standard statistical learning tasks, only adjacent dependencies were learned. The very same pattern was observed when digits were replaced by syllables. These results show that the perfect symmetry found in Pacton and Perruchet was not due to the fact that the processing of digits is less sensitive to their distance than the processing of syllables, tones, or visual shapes used in most statistical learning tasks. Moreover, the present results, completed with a reanalysis of the data collected in Pacton and Perruchet (2008), demonstrate that participants are highly sensitive to violations involving the spacing between paired elements. Overall, these results are consistent with the Pacton and Perruchet's single-process account of adjacent and nonadjacent dependencies, in which the joint attentional processing of the two events is a necessary and sufficient condition for learning the relation between them, irrespective of their distance. However, this account should be completed to encompass the notion that the presence or absence of an intermediate event is an intrinsic component of the representation of an association.

How does Stroop interference change with practice? A reappraisal from the musical Stroop paradigm.

Most earlier studies investigating the evolution of the Stroop effect with the amount of reading practice have reported data consistent with an inverted U-shaped curve, whereby the Stroop effect appears early during reading acquisition, reaches a peak after 2 or 3 years of practice, and then continuously decreases until adulthood. The downward component of the curve suggests that skilled performers would be able to control their performance better than less-skilled performers. However, in these studies, the level of reading practice entirely coincides with age due to obvious practical and ethical constraints, and it is possible that the observed reduction in the Stroop interference is due to a growing ability of older children to inhibit nonrelevant information. In the present study, word reading, as source of interference, was replaced by note naming in musicians. The major advantage is that musical training can be easily decoupled from age. In 2 experiments exploiting the musical Stroop paradigm (Grégoire, Perruchet, & Poulin-Charronnat, 2013), we observed an early appearance of the interference effect, as reported for the color-word and picture-word Stroop tasks, but we did not replicate the inverted U-shaped curve. Experiment 2 revealed a linear and positive relation between the amplitude of the musical Stroop effect and the amount of musical practice across 5 years of musical training. These results suggest that reading practice in itself does not lead to increased control over reading and that the usual pattern of results is most likely due to the strong correlation between age and reading practice.

About the unidirectionality of interference: insight from the musical Stroop effect.

The asymmetry of interference in a Stroop task usually refers to the well-documented result that incongruent colour words slow colour naming (Stroop effect) but incongruent colours do not slow colour word reading (no reverse Stroop effect). A few other studies have suggested that, more generally, a reverse Stroop effect can be occasionally observed but at the expense of the Stroop effect itself, as if interference was inherently unidirectional, from the stronger to the weaker of the two competing processes. We describe here a situation conducive to a pervasive mutual interference effect. Musicians were exposed to congruent and incongruent note name/note position patterns, and they were asked either to read the word while ignoring the location of the note within the staff, or to name the note while ignoring the note name written inside the note picture. Most of the participants exhibited interference in the two tasks. Overall, this result pattern runs against the still prevalent model of the Stroop phenomenon [Cohen, J. D., Dunbar, K., & McClelland, J. L. (1990). On the control of automatic processes: A parallel distributed processing account of the Stroop effect. Psychological Review, 97(3), 332-361]. However, further analyses lend support to one of the key tenets of the model, namely that the pattern of interference depends on the relative strength of the two competing pathways. The reasons for the impressive differences between the results collected in the present study and in the standard colour-word (or picture-word) paradigms are also examined. We suggest that these differences reveal the importance of stimulus-response contingency in the formation of automatisms.

New evidence for chunk-based models in word segmentation.

There is large evidence that infants are able to exploit statistical cues to discover the words of their language. However, how they proceed to do so is the object of enduring debates. The prevalent position is that words are extracted from the prior computation of statistics, in particular the transitional probabilities between syllables. As an alternative, chunk-based models posit that the sensitivity to statistics results from other processes, whereby many potential chunks are considered as candidate words, then selected as a function of their relevance. These two classes of models have proven to be difficult to dissociate. We propose here a procedure, which leads to contrasted predictions regarding the influence of a first language, L1, on the segmentation of a second language, L2. Simulations run with PARSER (Perruchet & Vinter, 1998), a chunk-based model, predict that when the words of L1 become word-external transitions of L2, learning of L2 should be depleted until reaching below chance level, at least before extensive exposure to L2 reverses the effect. In the same condition, a transitional-probability based model predicts above-chance performance whatever the duration of exposure to L2. PARSER's predictions were confirmed by experimental data: Performance on a two-alternative forced choice test between words and part-words from L2 was significantly below chance even though part-words were less cohesive in terms of transitional probabilities than words.

Is the musical stroop effect able to keep its promises? A reply to Akiva-Kabiri and Henik (2014), Gast (2014), Moeller and Frings (2014), and Zakay (2014).

Grégoire, Perruchet, and Poulin-Charronnat (2013) claimed that the Musical Stroop task, which reveals the automaticity of note naming in musician experts, provides a new tool for studying the development of automatisms through extensive training in natural settings. Many of the criticisms presented in the four commentaries published in this issue appear to be based on a misunderstanding of our procedure, or questionable postulates. We maintain that the Musical Stroop Effect offers promising possibilities for further research on automaticity, with the main proviso that the current procedure makes it difficult to tease apart facilitation and interference.

The musical Stroop effect. Opening a new avenue to research on automatisms.

The usual color-word Stroop task, as well as most other Stroop-like paradigms, has provided invaluable information on the automaticity of word reading. However, investigating automaticity through reading alone has inherent limitations. This study explored whether a Stroop-like effect could be obtained by replacing word reading with note naming in musicians. Note naming shares with word reading the crucial advantage of being intensively practiced over years by musicians, hence allowing to investigate levels of automatism that are out of reach of laboratory settings. But the situation provides much greater flexibility in manipulating practice. For instance, even though training in musical notation is often conducted in parallel with the acquisition of literacy skills during childhood, many exceptions make that it can be easily decoupled from age. Supporting the possibility of exploiting note naming as a new tool for investigating automatisms, musicians asked to process note names written inside note pictures in incongruent positions on a staff were significantly slowed down in both a go/no-go task (Experiment 1) and a verbal task (Experiment 2) with regard to a condition in which note names were printed inside note pictures in congruent positions.

Challenging prior evidence for a shared syntactic processor for language and music.

A theoretical landmark in the growing literature comparing language and music is the shared syntactic integration resource hypothesis (SSIRH; e.g., Patel, 2008), which posits that the successful processing of linguistic and musical materials relies, at least partially, on the mastery of a common syntactic processor. Supporting the SSIRH, Slevc, Rosenberg, and Patel (Psychonomic Bulletin & Review 16(2):374-381, 2009) recently reported data showing enhanced syntactic garden path effects when the sentences were paired with syntactically unexpected chords, whereas the musical manipulation had no reliable effect on the processing of semantic violations. The present experiment replicated Slevc et al.'s (2009) procedure, except that syntactic garden paths were replaced with semantic garden paths. We observed the very same interactive pattern of results. These findings suggest that the element underpinning interactions is the garden path configuration, rather than the implication of an alleged syntactic module. We suggest that a different amount of attentional resources is recruited to process each type of linguistic manipulations, hence modulating the resources left available for the processing of music and, consequently, the effects of musical violations.

Centre-embedded structures are a by-product of associative learning and working memory constraints: evidence from baboons (Papio Papio).

Influential theories have claimed that the ability for recursion forms the computational core of human language faculty distinguishing our communication system from that of other animals (Hauser, Chomsky, & Fitch, 2002). In the present study, we consider an alternative view on recursion by studying the contribution of associative and working memory processes. After an intensive paired-associate training with visual shapes, we observed that baboons spontaneously ordered their responses in keeping with a recursive, centre-embedded structure. This result suggests that the human ability for recursion might partly if not entirely originate from fundamental processing constraints already present in nonhuman primates and that the critical distinction between animal communication and human language should more likely be found in working memory capacities than in an ability to produce recursive structures per se.

Exploiting multiple sources of information in learning an artificial language: human data and modeling.

This study investigates the joint influences of three factors on the discovery of new word-like units in a continuous artificial speech stream: the statistical structure of the ongoing input, the initial word-likeness of parts of the speech flow, and the contextual information provided by the earlier emergence of other word-like units. Results of an experiment conducted with adult participants show that these sources of information have strong and interactive influences on word discovery. The authors then examine the ability of different models of word segmentation to account for these results. PARSER (Perruchet & Vinter, 1998) is compared to the view that word segmentation relies on the exploitation of transitional probabilities between successive syllables, and with the models based on the Minimum Description Length principle, such as INCDROP. The authors submit arguments suggesting that PARSER has the advantage of accounting for the whole pattern of data without ad-hoc modifications, while relying exclusively on general-purpose learning principles. This study strengthens the growing notion that nonspecific cognitive processes, mainly based on associative learning and memory principles, are able to account for a larger part of early language acquisition than previously assumed.

The influence of temporal factors on automatic priming and conscious expectancy in a simple reaction time task.

In a previous study, we reported a dissociation between subjective expectancy and motor behaviour in a simple associative learning task (Perruchet, Cleeremans, & Destrebecqz, 2006). According to previous conditioning studies (Clark, Manns, & Squire, 2001), this dissociation is observed when the to-be-associated events coterminate and thus overlap in time (a training regimen called delay conditioning), but not when they are separated by a temporal delay (trace conditioning). In this latter situation indeed, there tends to be a direct relationship between subjective expectancy and behaviour. In this study, we further investigated this issue in a series of experiments where conscious and unconscious components of performance were pitted against each other. In Experiments 1-3, participants performed a simple reaction time task in which a preparatory signal (a tone) either overlapped with or terminated earlier than the imperative stimulus (a visual target presented in 50% of the trials). After each response, participants also had to state how much they expected the imperative stimulus to be displayed on the next trial. Results indicate that reaction times tend to decrease when the tone is consistently followed by the visual target across successive trials, whereas conscious expectancy for the target decreases at the same time. Importantly, we systematically found that the temporal relationship between the tone and the target failed to influence performance. In a fourth experiment, we examined whether these results extend to a two-choice reaction time task. To our surprise, we observed a direct relationship between subjective expectancies and reaction time in that situation. We nevertheless observed that the introduction of a delay between the tone and the target had, once again, no effect on performance.

Generating constrained randomized sequences: item frequency matters.

All experimental psychologists understand the importance of randomizing lists of items. However, randomization is generally constrained, and these constraints-in particular, not allowing immediately repeated items-which are designed to eliminate particular biases, frequently engender others. We describe a simple Monte Carlo randomization technique that solves a number of these problems. However, in many experimental settings, we are concerned not only with the number and distribution of items but also with the number and distribution of transitions between items. The algorithm mentioned above provides no control over this. We therefore introduce a simple technique that uses transition tables for generating correctly randomized sequences. We present an analytic method of producing item-pair frequency tables and item-pair transitional probability tables when immediate repetitions are not allowed. We illustrate these difficulties and how to overcome them, with reference to a classic article on word segmentation in infants. Finally, we provide free access to an Excel file that allows users to generate transition tables with up to 10 different item types, as well as to generate appropriately distributed randomized sequences of any length without immediately repeated elements. This file is freely available from http://leadserv.u-bourgogne.fr/IMG/xls/TransitionMatrix.xls.

Implicit learning of nonlocal musical rules: a comment on Kuhn and Dienes (2005).

In a recent study, G. Kuhn and Z. Dienes (2005) reported that participants previously exposed to a set of musical tunes generated by a biconditional grammar subsequently preferred new tunes that respected the grammar over new ungrammatical tunes. Because the study and test tunes did not share any chunks of adjacent intervals, this result may be construed as straightforward evidence for the implicit learning of a structure that was only governed by nonlocal dependency rules. It is shown here that the grammar modified the statistical distribution of perceptually salient musical events, such as the probability that tunes covered an entire octave. When the influence of these confounds was removed, the effect of grammaticality disappeared.

Does unconscious thought improve complex decision making?

In a recent study, Dijksterhuis et al. (Science 311:1005, 2006) reported that participants were better at solving complex decisions after a period of unconscious thought relative to a period of conscious thought. They interpreted their results as an existence proof of powerful unconscious deliberation mechanisms. In the present report, we used a similar experimental design with an additional control, immediate condition, and we observed that participants produced as good (and even descriptively better) decisions in this condition than in the "unconscious" one, hence challenging the initial interpretation of the authors. However, we still obtained lower performances in the "conscious" relative to the "immediate" condition, suggesting that the initial result of Dijksterhuis et al. was not due to the action of powerful unconscious thought processes, but to the apparent disadvantage of further conscious processing. We provide an explanation for this observation on the basis of current models of decision making. It is finally concluded that the benefit of unconscious thought in complex decision making is still a controversial issue that should be considered cautiously.

A role for backward transitional probabilities in word segmentation?

A number of studies have shown that people exploit transitional probabilities between successive syllables to segment a stream of artificial continuous speech into words. It is often assumed that what is actually exploited are the forward transitional probabilities (given XY, the probability that X will be followed by Y), even though the backward transitional probabilities (the probability that Y has been preceded by X) were equally informative about word structure in the languages involved in those studies. In two experiments, we showed that participants were able to learn the words from an artificial speech stream when the only available cues were the backward transitional probabilities. Learning is as good under those conditions as when the only available cues are the forward transitional probabilities. Implications for some current models of word segmentation, particularly the simple recurrent networks and PARSER models, are discussed.