Psychology of Learning and Motivation (eBook)

eBook Download: EPUB

2015 | 1. Auflage
250 Seiten
Elsevier Science (Verlag)
978-0-12-802471-3 (ISBN)

Infos

Psychology of Learning and Motivation publishes empirical and theoretical contributions in cognitive and experimental psychology, ranging from classical and instrumental conditioning to complex learning and problem solving. Each chapter thoughtfully integrates the writings of leading contributors, who present and discuss significant bodies of research relevant to their discipline. Volume 62 includes chapters on such varied topics as automatic logic and effortful beliefs, complex learning and development, bias detection and heuristics thinking, perceiving scale in real and virtual environments, using multidimensional encoding and retrieval contexts to enhance our understanding of source memory, causes and consequences of forgetting in thinking and remembering and people as contexts in conversation.

• Volume 62 of the highly regarded Psychology of Learning and Motivation series
• An essential reference for researchers and academics in cognitive science
• Relevant to both applied concerns and basic research

Textauszug

Chapter Two

Dual Processes and the Interplay between Knowledge and Structure: A New Parallel Processing Model

Simon J. Handley1 and Dries Trippas     Cognition Institute, School of Psychology, Plymouth University, Plymouth, UK
1 Corresponding author: E-mail: [e-mail] 

Abstract

How do reasoners resolve a conflict between two competing responses, one cued by beliefs and knowledge, and one based upon the problem's underlying logic and structure? The literature suggests that such conflicts are routinely resolved in favor of a belief-based heuristic (Type 1) response that is generated autonomously and by default. Given sufficient effort, time, and motivation, reasoners can and sometimes do engage in more deliberative (Type 2) processing and intervene on default responses, generating responses based upon the underlying structure of the problem. Such default interventionist accounts of reasoning biases have become increasingly popular in the reasoning and judgment domains. In this chapter we review recent evidence which suggests that reasoners show intuitive sensitivity to logical structure, which can interfere with belief or knowledge judgments. We further show that the tendency to respond on the basis of beliefs can be slow and effortful, can increase with development, and often depends upon working memory. We evaluate the implications of these findings for dual process accounts and conclude that existing dual process frameworks cannot account for the evidence. Finally, we sketch the principles of a new model which assumes that both structure and knowledge make intuitive and deliberative contributions to responses on reasoning and judgment tasks.

Keywords

Reasoning; dual processes; beliefs; heuristics; biases; intuitive logic

1. Introduction

Robin is a make-up artist with a fascination for dancing who enjoys drinking cosmopolitans. When asked to assess Robin's gender, based on our knowledge of stereotypes, it is tempting to assume he or she is female. Now assume that Robin is a randomly sampled individual from a population in which less than 1% of the attendees are women (e.g., a Star Trek Convention). Is it still likely that Robin is female? The random nature of the sampling procedure coupled with the low base rate of women in the scrutinized population makes it unlikely that Robin is female. Nevertheless, research shows that people very often neglect the base rate, responding on the basis of the stereotype instead (Kahneman, 2011). This example is a classic illustration of the presence of conflicting responses, one of which is based upon readily available knowledge and a second that depends upon deeper processing.

Dual process theory (DPT) is a metatheoretical framework which posits that the conflict between two types of cognitive processes can explain the majority of human decision-making patterns (Evans & Stanovich, 2013). Type 1 (T1) processes are effortless, high capacity, autonomous, and as a consequence, quick. Consider for example the following expression:

On a sunny day the sky is colored ____

According to DPT, T1 processing is responsible for the immediate generation of the response “blue.” More importantly, it is difficult to suppress this instantaneous response. Type 2 (T2) processes are effortful, low capacity, reliant upon working memory resources, and therefore slow. Once again, as an example, consider the following expression (Thompson, 2013):

53 × 96 = ____

The solution (5088) can be found without the use of a calculator, but doing so will require some level of effort—i.e.,T2 processing. These processes are thought to play an important role in the majority of our daily behaviors. Stereotyping and substance abuse, for instance, are behaviors characterized by T1–T2 conflicts where T1 processes take the upper hand. In contrast, correct logical deductions and delayed gratification are behaviors in which this alluring T1 response was avoided in favor of the more effortful T2 response. Base-rate neglect is one of the many paradigms used to study dual processing.

In a typical base-rate neglect vignette (cf., above), a T1–T2 conflict is created by presenting people with a stereotype which contradicts the base rate (e.g., De Neys & Glumicic, 2008). According to DPT, T1 processes cue the stereotype-based response, whereas T2 processes are necessary to correctly provide the normative base-rate-based response. In short, stereotypes are cued by T1 processing and the correct use of base rates requires T2 processing. This is known as the default interventionist (DI) account of DPT (Evans, 2007, 2010).

The DI account more generally posits that in the vast majority of reasoning and judgment, T1 processing will rapidly provide a default response, usually on the basis of prior knowledge. If a T1–T2 conflict is detected, and if certain conditions are met, people can inhibit this intuition using additional T2 processing–required to make use of the formal problem structure. This raises an important question, however: if T2 processing cannot and does not start until a T1–T2 conflict is detected, how is the conflict detected in the first place?

Sloman (1996; see also De Neys, 2012; Handley, Newstead, & Trippas, 2011) introduced a parallel competitive (PC) dual process account which can cope with this issue. According to the PC account, T1 and T2 processing occur simultaneously and start at the problem onset. People can thus become aware of a potential conflict, but not always, given that T1 processes are much more dominant due to their autonomy and speed. Nevertheless, a large body of recent research has been interpreted as evidence for the DI account. Our aim in this chapter is to argue against this interpretation, instead proposing an alternative PC characterization of DPT. More importantly, we will argue that the intuitively appealing associations between T1 and knowledge, and T2 and structure, are fallacious. Instead, we will argue for a PC DPT model in which structural problem features and prior knowledge are activated simultaneously, both potentially reliant on T1 and T2 processing, depending on various determinants such as salience, fluency, and complexity.

This chapter is structured as follows. First, we review the most relevant modern studies typically interpreted as evidence in favor of default interventionism. Next, we delve into the concepts of conflict and conflict detection in more detail. We move on to demonstrate in two separate sections that structural problem features can be processed quickly, and that knowledge use is sometimes effortful and slow. In the general discussion we discuss the implications for DPTs, and we argue that default interventionism as a dual process subtheory is untenable.

2. Dual Processes and DI Accounts

In the study of human reasoning and decision making, three paradigms are often used to test the predictions drawn from DPT: the conjunction fallacy, belief bias, and base-rate neglect. A description of each can be found in Table 1.

These experimental tasks have in common that a heuristic response based on prior knowledge and cued by T1 processes is in conflict with an analytical response which requires the use of the problem structure, and thus T2 processing. Remember that according to the DI interpretation of DPT, T1 processing precedes T2 processing because it is quick and effortless, in contrast to T2 processing, which is slow due to its effortful reliance on working memory. A straightforward prediction is that earlier responses should be predominantly based on knowledge, with later responses being based more on structure.

Evans and Curtis-Holmes (2005) tested this prediction by manipulating the response deadline in a belief bias paradigm. Participants were presented with reasoning problems in a self-paced or time-limited response condition. Consistent with the predictions made by the DI DPT account, the time-pressured group showed significantly increased levels of belief bias as well as decreased levels of logical responding, compared to the self-paced control group. De Neys (2006) focused on the working memory component of the prediction. According to the DI account, a working memory load should interfere with the ability to engage T2 processing (logical structure). In contrast, belief-based responding should remain unaffected even in the face of insufficient working memory resources, given its T1 nature. The participants performed a secondary task designed to put a load on working memory to some degree (high, low, or no load). The results demonstrated that there was no effect of working memory load on belief-based responding. Logic-based responding, however, significantly decreased as working memory load increased.

 

Table 1

Three typical paradigms used in the study of dual processes in reasoning

Conjunction fallacy task:

Sarah is 12 years of age. She is very talkative and sociable. She goes to drama classes and is learning to play the guitar. She wants to be a pop singer or an actress.

Which one of the following statements is more likely?

1. Sarah likes to cook*

2. Sarah likes to cook and she collects pop magazines

Belief bias task:

All living things need water

Roses need water

Therefore, roses are living things

1. The conclusion is...

Produkt-Details

• Erscheint lt. Verlag: 21.1.2015
• Mitarbeit: Herausgeber (Serie): Brian H. Ross
• Sprache: englisch
• Themenwelt: Geisteswissenschaften ► Psychologie ► Allgemeine Psychologie
• Themenwelt: Geisteswissenschaften ► Psychologie ► Biopsychologie / Neurowissenschaften
• Themenwelt: Geisteswissenschaften ► Psychologie ► Pädagogische Psychologie
• Themenwelt: Geisteswissenschaften ► Psychologie ► Verhaltenstherapie
• Themenwelt: Naturwissenschaften ► Biologie ► Humanbiologie
• Themenwelt: Naturwissenschaften ► Biologie ► Zoologie
• ISBN-10: 0-12-802471-2 / 0128024712
• ISBN-13: 978-0-12-802471-3 / 9780128024713