Dual Processes, Development, and Scalar Implicature
Year: 2006
Type: conference-paper
Abstract: Dual Processes, Development, and Scalar Implicature Susan Scrafton ([email protected]) Child Development Research Unit, Wolfson Research Institute, Queen’s Campus, Durham University, Stockton-on-Tees TS17 6BH United Kingdom Aidan Feeney ([email protected])) Applied Psychology, Queen’s Campus, Durham University Stockton-on-Tees TS17 6BH United Kingdom process. However, we will first consider the distinction in more detail and review the very limited developmental literature that it has motivated. Evidence for the existence of two types of thinking comes from a range of sources using a variety of methodologies. One important piece of evidence concerns people’s ability to resist background beliefs when engaged in deductive reasoning. In the belief bias paradigm (Evans, Barston & Pollard, 1983) participants are presented with syllogisms where the validity of the conclusions and the believability of the premises have been crossed in a fully factorial design. Although people tend to accept more valid than invalid conclusions, they also tend to accept more conclusions that are believable than unbelievable. This finding is taken as evidence for the co-existence of belief-based heuristic processes, and separate analytic processes capable of decontextualising the information in the problem. Further studies of belief bias have shown that it is more common in older people (Gilinsky & Judd, 1994) and under conditions of speeded response (Evans & Curtis-Holmes, 2005). The tendency to respond in accord with logical validity on the belief bias task is also known to be associated with IQ (Stanovich & West, 1998). These findings support the claim that there are separate processes for reasoning, one of which is fast, computationally inexpensive and context-driven, whilst the other is slow, computationally expensive, heavily dependent on working memory and responsible for dexcontextualising the problem. Furthermore, neuropsychological studies have shown a within-subject anatomical disassociation between areas associated with belief-based responding and those associated with logical responding (Goel & Dolan, 2004). Abstract Dual process theories suggest that our thinking is determined by separate, but interacting, heuristic and analytic processes. We describe an experiment designed to investigate the development of these processes. We examined the extent to which groups of 6, 9, 12, and 15 year-old participants and a group of adults were sensitive to scalar implicature in pragmatically enriched and pragmatically impoverished contexts. For younger participants we find high rates of pragmatic responding in enriched contexts and high rates of logical responding in impoverished contexts. Early adolescents displayed universally pragmatic responding whilst logical responding re-emerged in older adolescents and adults. These results suggest that heuristic processes develop earlier than analytic processes but that they are not well developed until early adolescence. Keywords: Reasoning, pragmatics, heuristics, context Introduction Cognitive development has traditionally been viewed as unidirectional, with the simple, inefficient and intuitive thought processes of childhood gradually replaced by more complex, efficient, systematic and analytical processes. However, this view that simpler processes are supplanted by more complex ones is challenged by dual-process theorists who suggest that different kinds of thinking may interact (see Evans & Over, 2004; Sloman, 1996; 2002; Stanovich, 1999; Stanovich & West, 2000). According to dual-process theories (for recent reviews see Evans, 2003; Osman, 2004) there are two separate but interacting processes for reasoning. Heuristic processes are believed to be fast, parallel, automatic, and unconscious in nature. They are often described as being associative, as they result in pragmatic responding that is often driven by context. Importantly, they are also cognitively undemanding and hence, may emerge relatively early in development. Analytic processes, on the other hand, are conscious, controlled, constrained by working memory capacity and slower than heuristic processes. In addition, they are held to be responsible for decontextualised thinking (Stanovich, 1999), which underlies our ability to reason or decide independently of context. The main aim of this paper will be to consider the developmental trajectory of the two types of Developmental Evidence Dual process theories are based, by and large, on adult data and have not been widely applied to reasoning in children and adolescents. A notable exception is Klaczynski’s work on adolescent thinking (Klaczynski, 2001; Klaczynski & Cottrell, 2004). Klaczynski (2001) examined the gap between young and middle adolescents’ performance on reasoning tasks and traditional standards of reasoning. He showed that rather than being random errors, non-normative responses were positively correlated, normative responses were positively correlated, and the two forms of response
Authors: Susan Scrafton, Aidan Feeney
Institution: Durham University
Language: English
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2600445554