Expertise Novices and experts Expertise and perception Expertise and memory Expertise and judgment Expertise and domain-specificity

Download Expertise Novices and experts Expertise and perception Expertise and memory Expertise and judgment Expertise and domain-specificity

Post on 03-Jan-2016




1 download

Embed Size (px)


<ul><li><p>ExpertiseNovices and expertsExpertise and perceptionExpertise and memoryExpertise and judgmentExpertise and domain-specificity</p></li><li><p>What is an expert?Experts are not just smart.You would not go to Einstein for a toothache.Experts know about a particular domain.</p></li><li><p>Questions about expertiseWhat does it mean to be an expert?Do experts see the world differently from novices?Do experts remember situations differently?How do you become an expert?Is any expertise domain-general?</p></li><li><p>What is an expert?Recognizes the problem in a new situationAble to diagnose new problemsKnows the answer to many problemsHas lots of prior experienceNeed not start from scratch solving problemsKnows how to solve new problemsSet of skills for solving unseen problemsFor ill-formed problems, finds operators that can apply to new problems.</p></li><li><p>Expertise and perceptionExperts may see the world differentlyThis is an MRI of a torn tendon.People must learn the relevant perceptual features in their domain.Studies of RadiologistsMedical students look at different features than experts.</p></li><li><p>Expertise and perceptionMany domains provide this type of expertise</p></li><li><p>Why does perception change?Return to the issue of constraintsThe cognitive system wants to automatize as much as possibleFrees up working memory resources for hard cognitive workIf perception can provide information useful for solving problems, tasks become easier.Perception limits the possibilities consideredGenerally a good thingExperts may miss novel features</p></li><li><p>Expertise and memoryThe way experts perceive a situation influences what they recall.DeGroot; Chase &amp; SimonChess experts memory for real board positions better than novices.Experts could recreate a board position faster than novices Their chunks were meaningful.Similar effects have been found in many other domains like sports.</p></li><li><p>Expert perception and categoriesBrooks and colleaguesDoctors combine perception with other diagnosis informationMay miss very obvious features without a case history or with a misleading case historyMay be a confirmation bias.A patient with jaundice</p></li><li><p>Expertise and judgmentAre expert judgments always better?Particularly in fuzzy domainsStock marketTwo ways to answer that questionCompare experts to statistical modelsCompare experts to novicesHow good are expert judgments?Statistical models generally outpredict expertsExperts pay too much attention to rare eventsExperts in combination with statistical models are better than either one alone.</p></li><li><p>Expert and novice judgmentIn some fuzzy domains, experts are no more accurate than novicesInterpreting Rorschach testsExperts are (usually) better calibrated than novicesTheir confidence in judgments matches their accuracyExperts may sometimes be more overconfidentWhy do experts perform poorly?Expert rules may overweight the wrong cuesExperts may only search for particular cues in particular contexts</p></li><li><p>Becoming an expertPractice, practice, practiceWhat kind of practice is needed?Development of perceptual skillsDevelopment of motor skillsFamiliarity with problems in a domainDiagnosing problemsFixing problemsSome problems only arise in rare cases, so expertise may take a while to developLearning what you do and do not knowExpertise: 10 years to develop (Gardner)</p></li><li><p>Domain specificity of expertiseExperts are experts in a domainAn expert in science need not be an expert in other academic disciplinesAn expert in once science need not be an expert in another scienceAttempts to teach general problem solving have been unsuccessfulPolya and heuristicsHeuristics are often too general to be usedFind a similar problemHow do you know what is similar without expertise?</p></li><li><p>This is an old ideaThorndike &amp; Woodworth (1901)Examined relationships among mental tasksImproving in one task only improved performance in tasks that had common componentsSkill in math improves skill in scienceSkill in math is unrelated to skill in languagesMore recent models (e.g., Anderson)ACT modelCognitive skills involve components (called productions)One skill improves another when they use the same productions.</p></li><li><p>SummaryExpertise takes time to developExperts see the world differently from novicesInfluences their memoryExperts have skills that novices do notExpertise is domain specificAttempts to teach general problem solving strategies generally fail.</p></li></ul>