Implicit learning and statistical learning: One phenomenon, two approaches

Perruchet, P. & Pacton, S. (2006) Implicit learning and statistical learning: One phenomenon, two approaches. Trends in Cognitive Sciences, 10(5).

People have a distinct ability to detect patterns and this ability has been explained using two theories: implicit learning and statistical learning. Implicit learning theory suggests that after sufficient exposure to a pattern one automatically deduces a rule regarding the formation of that pattern which can then be accessed consciously. In contrast, statistical learning suggests no explicit rule learning is necessary and rather, that our brains perform statistical computations to predict the likelihood that a certain stimulus will occur given the occurrence of another stimulus. These theories have huge implications in the world of language acquisition, where words and grammatical structures must be intuited from continuous sequences of syllables.

In the present paper, Perruchet and Pacton (2006) make it clear that while the origins of implicit and statistical learning theories are vastly different, they have converged to a single goal: to explain general learning through domain-general processing. Further, Perruchet and Pacton (2006) suggest two explanations which attempt to integrate the two theories.

The first explanation suggests all levels of learning are done with chunks based on the frequency with which they occur. This model claims that statistical patterns, which statistical learning theory predicts are used to acquire information about a pattern, are only a byproduct of the frequency with which different chunks occur. The plausibility of this theory is supported by competitive chunking models like PARSER (Perruchet & Vinter, 1998, as cited in Perruchet and Pacton) which posits that chunks which maintain cognitive representation are those which are most frequent.

The second theory suggests that statistical computations are carried out to form the chunks and then these chunks are used in a competitive model. As Perruchet and Pacton explain, this explanation has the benefit of explaining why the stimulus preceding and following a string can impact the way it is remembered, a fact not well explained by competitive chunking models of learning like PARSER. The paper concludes that there is insufficient evidence to support the superiority of either the pure chunking or the statistical chunking model of learning. It is suggested the distinction between statistical and implicit learning is important in determining the involvement of consciousness in learning, as implicit learning theory suggests chunks are manipulated consciously whereas statistical learning theory would indicate that chunk knowledge would be held in the unconscious.

Blogger; Braxton Murphy is an undergraduate research student working under the supervision of Dr. Lisa Archibald

Design-Based Research: Putting a Stake in the Ground

Barab, S., &Squire, K. (2004). Design-based research: Putting a stake in the ground. The Journal of the Learning Sciences, 13(1), 1-14.

Learning sciences refers to the study of learning and instructional methodologies. One approach to this work is design-based research, the goal of which is to create new theories, and practices impacting learning and teaching in a real-life setting. In this approach, researchers systematically asses the impact of changes to the learning context. Barab & Squire (2004) outline seven differences between design-based research and traditional methodology. Some of these distinctions include: the location of research, complexity of variables, focus of research, and role of participants. These differences emphasize that design-research often occurs in a real-life setting, that measurement is challenging due to the continuously changing context, and that some researchers may be both designing and participating in the study.  A large distinction between traditional research and design-research is that design-research requires change at a local level and this change is used as evidence to support the theory behind the design.

The authors consider how to measure overall change in this approach and they pose the question - what counts as credible research? In a design-based research approach, the terms trustworthiness, credibility, and usefulness capture the study’s reliability, validity, and generalizability/external validity. Some critics of design-based research believe problems arise when the effectiveness of design-based research is evaluated. This is because it is the researcher who is determining the effectiveness that is also the designer and participated in the interactions assessed. However, other researchers argue that design-research can be adaptable to uniquely fit a local dynamic, and thereby the goal is to develop flexible theories applicable to the current and new contexts.  

Creating design-research that is usable and sustainable when implemented in real-world contexts may be an important facilitator of researcher-practitioner collaboration. Design-researchers work together to provide credible, trustworthy, and useful evaluation of instructional methods in a real-world environment.

Blogger: Meghan Vollebregt is a student in the combined SLP MSc/PhD program working under the supervision of Lisa Archibald.

The role of language in mathematical development: Evidence from children with specific language impairments

Donlan, C., Cowan, R., Newton, E.J., Lloyd, D. (2007). The role of language in mathematical development: Evidence from children with specific language impairments. Cognition,103. 23-33.

Although language and mathematics are distinct skills, some studies suggest that these two cognitive processes are related. One view suggests that language plays a bootstrapping role in numerical cognition, in which development of number concepts is dependent on number-related language experience¹. An alternate view proposes that number concept development is independent of number word knowledge². Children with specific language impairment (SLI) have impairments in receptive and expressive language but have non-verbal cognitive abilities within the average range. By studying numerical cognition in children with SLI, the present study aims to examine whether language impairments contribute to difficulty in mathematical cognition. Specifically, the paper differentiates between procedural mathematical knowledge (counting and basic calculations) and conceptual mathematical knowledge (understanding of place value and arithmetic principles) to examine the relative contributions of language and non-verbal abilities to these aspects of numerical cognition.

The present study examined three groups of children: children with SLI, typically developing children matched to the SLI group based on age, and typically developing children matched to the SLI group based on language comprehension. These participants all completed the following tasks: counting aloud, simple calculations, multi-digit magnitude comparison, and arithmetic problems using unfamiliar symbols. The counting and calculations tasks were categorized as procedural tasks, while the magnitude comparison and arithmetic problems tasks were categorized as conceptual as they were designed to measure understanding of place value principles and arithmetic principles.

Results demonstrated that on the counting and calculation tasks, the SLI group performed similarly to the language controls and more poorly than the age controls. On the magnitude comparison tasks, the age controls outperformed the SLI group and the SLI group outperformed the language controls. On the arithmetic principles task, the SLI and age controls performed similarly and both outperformed the language controls. Counting skills were a significant predictor of calculation and magnitude comparison performance. These findings suggest that the children with SLI are able to achieve conceptual understanding of mathematical principles, but that their language weaknesses may contribute to difficulty developing procedural mathematical skills. The authors suggest that conceptual understanding of arithmetic may be supported by a system separate from language but that language may support learning of the counting sequence, which, in turn, supports understanding of calculation and number notation.

1. Carey, S. (2004). Bootstrapping and the origin of concepts. Daedalus, 133, 59–68.

2. Gelman, R., & Butterworth, B. (2005). Number and language: How are they related? Trends in Cognitive Sciences 9, 6–10.

Blogger: Alex Cross is completing a combined MClSc and PhD in speech language pathology. Her work focusing on reading will be part of both the Language and Working Memory and the Language, Reading, and Cognitive Neuroscience labs.