Suppr超能文献

疾病表征中的信息处理:来自联想学习框架的启示

Information processing in illness representation: Implications from an associative-learning framework.

作者信息

Lowe Rob, Norman Paul

机构信息

Psychology Department, Swansea University.

Department of Psychology, University of Sheffield.

出版信息

Health Psychol. 2017 Mar;36(3):280-290. doi: 10.1037/hea0000457. Epub 2016 Dec 8.

Abstract

OBJECTIVE

The common-sense model (Leventhal, Meyer, & Nerenz, 1980) outlines how illness representations are important for understanding adjustment to health threats. However, psychological processes giving rise to these representations are little understood. To address this, an associative-learning framework was used to model low-level process mechanics of illness representation and coping-related decision making.

METHOD

Associative learning was modeled within a connectionist network simulation. Two types of information were paired: Illness identities (indigestion, heart attack, cancer) were paired with illness-belief profiles (cause, timeline, consequences, control/cure), and specific illness beliefs were paired with coping procedures (family doctor, emergency services, self-treatment). To emulate past experience, the network was trained with these pairings. As an analogue of a current illness event, the trained network was exposed to partial information (illness identity or select representation beliefs) and its response recorded.

RESULTS

The network (a) produced the appropriate representation profile (beliefs) for a given illness identity, (b) prioritized expected coping procedures, and (c) highlighted circumstances in which activated representation profiles could include self-generated or counterfactual beliefs.

CONCLUSIONS

Encoding and activation of illness beliefs can occur spontaneously and automatically; conventional questionnaire measurement may be insensitive to these automatic representations. Furthermore, illness representations may comprise a coherent set of nonindependent beliefs (a schema) rather than a collective of independent beliefs. Incoming information may generate a "tipping point," dramatically changing the active schema as a new illness-knowledge set is invoked. Finally, automatic activation of well-learned information can lead to the erroneous interpretation of illness events, with implications for [inappropriate] coping efforts. (PsycINFO Database Record

摘要

目的

常识模型(Leventhal、Meyer和Nerenz,1980)概述了疾病表征对于理解应对健康威胁的重要性。然而,产生这些表征的心理过程却鲜为人知。为了解决这个问题,一个联想学习框架被用于模拟疾病表征和应对相关决策的低层次过程机制。

方法

在一个联结主义网络模拟中对联想学习进行建模。两种类型的信息被配对:疾病身份(消化不良、心脏病发作、癌症)与疾病信念概况(病因、病程、后果、控制/治愈)配对,特定的疾病信念与应对程序(家庭医生、急救服务、自我治疗)配对。为了模拟过去的经验,网络用这些配对进行训练。作为当前疾病事件的模拟,训练后的网络被输入部分信息(疾病身份或选定的表征信念)并记录其反应。

结果

该网络(a)为给定的疾病身份生成了适当的表征概况(信念),(b)对预期的应对程序进行了优先级排序,(c)突出了激活的表征概况可能包括自我生成或反事实信念的情况。

结论

疾病信念的编码和激活可以自发且自动地发生;传统的问卷调查可能对这些自动表征不敏感。此外,疾病表征可能由一组连贯的非独立信念(一种图式)组成,而不是独立信念的集合。传入的信息可能会产生一个“临界点”,当调用新的疾病知识集时,会显著改变活跃的图式。最后,对充分学习的信息的自动激活可能会导致对疾病事件的错误解读,从而影响[不适当的]应对努力。(PsycINFO数据库记录)

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验