Advanced

Metacognition and Confidence: Mood, Individual Differences, Developmental and Social Aspects

Jonsson, Anna-Carin LU (2005) In Scandinavian Journal of Psychology, Personality and Individual Differences, Applied Cognitive Psychology
Abstract (Swedish)
Popular Abstract in Swedish

Denna avhandling undersöker realismen i människors konfidensbedömningar. En konfidens- bedömning sker när en person uttalar hur säker han/hon är på att en proposition han/hon har framfört är korrekt. Med realism menas hur väl personens konfidensbedömning matchar korrektheten i uttalandet. Många olika faktorer påverkar realismen i konfidensen och denna avhandling fokuserar på fyra faktorer; sinnesstämning, individuella skillnader, utvecklingspsykologiska och sociala aspekter.



Studie I undersökte om sinnesstämning påverkade den säkerhet människor har till sitt episodiska minne. En positiv sinnesstämning inducerades i en grupp. Dock fanns ingen signifikant skillnad mellan de två... (More)
Popular Abstract in Swedish

Denna avhandling undersöker realismen i människors konfidensbedömningar. En konfidens- bedömning sker när en person uttalar hur säker han/hon är på att en proposition han/hon har framfört är korrekt. Med realism menas hur väl personens konfidensbedömning matchar korrektheten i uttalandet. Många olika faktorer påverkar realismen i konfidensen och denna avhandling fokuserar på fyra faktorer; sinnesstämning, individuella skillnader, utvecklingspsykologiska och sociala aspekter.



Studie I undersökte om sinnesstämning påverkade den säkerhet människor har till sitt episodiska minne. En positiv sinnesstämning inducerades i en grupp. Dock fanns ingen signifikant skillnad mellan de två gruppernas (positiv och neutral) konfidensbedömningar. Studiens slutsats var att moderata nivåer av en positiv sinnesstämning inte påverkar de aktuella metakognitiva bedömningarna. Frekvensbedömningarnas (ett mått på hur många frågor totalt som personen ifråga tror sig ha svarat rätt på) visade att deltagarna signifikant underskattade sin faktiska prestation, detta i kontrast till de specifika konfidensbedömningarna, där deltagarna överskattade sin faktiska prestation. I Studie II undersöktes stabiliteten och variabiliteten i realismen i deltagarnas konfidensbedömningar över tid, innehållsdomän, genus och kognitiv stil. Två test, ORD (ordkunskap) och DTK (logisk/spatial förmåga) från det svenska Högskoleprovet administrerades över tre tillfällen. Resultaten visade att det finns individuella stabila skillnader över tid och domän i deltagarnas konfidensbedömningar. Dock återfanns också skillnader mellan de två domänerna i deltagarnas konfidensbedömningar. Inga skillnader mellan genus kunde spåras och kognitiv stil (Need for Cognition) hade ingen effekt på deltagarnas konfidensbedömningar. Studie III undersökte barns (11-12 år) konfidensbedömningar till svar på frågor gällande deras minne (det episodiska minnet) av en kort videofilm. Fyra olika skalor att skatta konfidensbedömningar med testades; den numeriska skalan, bildskalan, linjeskalan och den verbala skalan. Vilken skala som användes påverkade inte barnens konfidensbedömningar. Barnen visade sig mer överkonfidenta i relation till sina svar än de vuxna som tidigare använt samma material i tidigare studier. Dessutom, barnens frekvensbedömningar (uppskattning av det totala antalet korrekt besvarade frågor) visade på att deltagarna överskattade sin faktiska prestation. Studie IV undersökte hur social återkoppling påverkade barns (12 år) konfidensbedömningar till svar på frågor gällande deras minne (det episodiska minnet) av en kort videofilm (samma material som användes i Studie III). Mellan-individ faktorn var relaterad till huruvida återkopplingen barnet fick kom från en lärare eller en kamrat. Inom-individ faktorn gällde olika typer av återkoppling; konfirmerande eller icke-konfirmerande återkoppling på såväl korrekta som inkorrekta svar. Resultatet visade att barnen (12 år) var överkonfidenta oavsett vilken typ av återkoppling de fick. Dock, konfirmerande återkoppling resulterade i sämre realism i barnens konfidensbedömningar än icke-konfirmerande återkoppling. Deltagarnas frekvensbedömningar (totalt antal korrekt besvarade frågor) visade återigen (se Studie III) på överskattning.



Resultatet av denna avhandling visar att ytterligare två faktorer, individuella skillnader och utvecklingspsykologiska aspekter, bör läggas till de fem; tid, återkoppling, kontext, motivation och erfarenhet inom domän, vilka introducerades av Allwood och Granhag (1999) som påverkande faktorer av människors konfidensbedömningar. (Less)
Abstract
This dissertation investigates the realism in people's confidence judgments. Study I investigated whether mood influences the confidence people have in their event memory. However, there were no significant differences in the realism of the participants' confidence judgments between conditions (elated and neutral). This study concluded that moderate levels of elation might not influence the realism in confidence judgments. In Study II the stability and variability in the realism of confidence judgments over time, content domain and gender were investigated. Two tests, WORD (word knowledge)and DTK (logical/spatial ability), from the Swedish Scholastic Aptitude Test were administered on three occasions. The results showed that there were... (More)
This dissertation investigates the realism in people's confidence judgments. Study I investigated whether mood influences the confidence people have in their event memory. However, there were no significant differences in the realism of the participants' confidence judgments between conditions (elated and neutral). This study concluded that moderate levels of elation might not influence the realism in confidence judgments. In Study II the stability and variability in the realism of confidence judgments over time, content domain and gender were investigated. Two tests, WORD (word knowledge)and DTK (logical/spatial ability), from the Swedish Scholastic Aptitude Test were administered on three occasions. The results showed that there were stable individual differences over time and domain in the participants' confidence judgments. No gender differences were found and no effect was found of the cognitive style Need-for-Cognition, NfC. Study III investigated children's (11-12 year old) confidence judgments of answers to questions on their episodic memories. Four different confidence scales were tested; the numeric scale, the picture scale, the line scale and the written scale. The four scales showed no difference for the children's confidence judgments. Further, the children were more overconfident than adults who had used the same material in recent studies. Also, the children's frequency judgments showed overestimation. Finally, Study IV investigated how social feedback influence children's (12 year old)confidence judgments (the same material as in Study III was used). The between-subject factor related to whether the feedback was given by a teacher or a classmate and the within-subject factor concerned type of feedback; confirmatory or disconfirmatory feedback on both correct and incorrect answers. The results showed that the children were overconfident independent of what kind of feedback they recieved and no effect of the source of the feedback was found. However, confirmatory feedback resulted in worse realism in the children's confidence judgments than disconfirmatory feedback. The children's frequency judgments again showed overestimation. In summary, the results in this thesis show that individual differences and developmental aspects should be added to the five factors that were argued by Allwood and Granhag (1999) to influence confidence judgments; time, feedback, contextual, motivational and experience within the domain. (Less)
Please use this url to cite or link to this publication:
author
supervisor
opponent
  • professor Biel, Anders, Psykologiska Institutionen, Göteborgs Universitet
organization
publishing date
type
Thesis
publication status
published
subject
keywords
Psykologi, Psychology, Developmental, Eyewitness memory., Individual Differences., Calibration., Realism in confidence judgments., Metacognition.
in
Scandinavian Journal of Psychology, Personality and Individual Differences, Applied Cognitive Psychology
pages
142 pages
publisher
Department of Psychology, Lund University
defense location
Kulturens auditorium, Lund
defense date
2005-01-28 13:15
ISBN
91-628-6387-8
language
English
LU publication?
yes
id
278ceb47-84d9-478a-ba97-17e084a1ef1f (old id 544153)
date added to LUP
2007-09-10 15:34:42
date last changed
2016-09-19 08:45:02
@misc{278ceb47-84d9-478a-ba97-17e084a1ef1f,
  abstract     = {This dissertation investigates the realism in people's confidence judgments. Study I investigated whether mood influences the confidence people have in their event memory. However, there were no significant differences in the realism of the participants' confidence judgments between conditions (elated and neutral). This study concluded that moderate levels of elation might not influence the realism in confidence judgments. In Study II the stability and variability in the realism of confidence judgments over time, content domain and gender were investigated. Two tests, WORD (word knowledge)and DTK (logical/spatial ability), from the Swedish Scholastic Aptitude Test were administered on three occasions. The results showed that there were stable individual differences over time and domain in the participants' confidence judgments. No gender differences were found and no effect was found of the cognitive style Need-for-Cognition, NfC. Study III investigated children's (11-12 year old) confidence judgments of answers to questions on their episodic memories. Four different confidence scales were tested; the numeric scale, the picture scale, the line scale and the written scale. The four scales showed no difference for the children's confidence judgments. Further, the children were more overconfident than adults who had used the same material in recent studies. Also, the children's frequency judgments showed overestimation. Finally, Study IV investigated how social feedback influence children's (12 year old)confidence judgments (the same material as in Study III was used). The between-subject factor related to whether the feedback was given by a teacher or a classmate and the within-subject factor concerned type of feedback; confirmatory or disconfirmatory feedback on both correct and incorrect answers. The results showed that the children were overconfident independent of what kind of feedback they recieved and no effect of the source of the feedback was found. However, confirmatory feedback resulted in worse realism in the children's confidence judgments than disconfirmatory feedback. The children's frequency judgments again showed overestimation. In summary, the results in this thesis show that individual differences and developmental aspects should be added to the five factors that were argued by Allwood and Granhag (1999) to influence confidence judgments; time, feedback, contextual, motivational and experience within the domain.},
  author       = {Jonsson, Anna-Carin},
  isbn         = {91-628-6387-8},
  keyword      = {Psykologi,Psychology,Developmental,Eyewitness memory.,Individual Differences.,Calibration.,Realism in confidence judgments.,Metacognition.},
  language     = {eng},
  pages        = {142},
  publisher    = {ARRAY(0x7baba58)},
  series       = {Scandinavian Journal of Psychology, Personality and Individual Differences, Applied Cognitive Psychology},
  title        = {Metacognition and Confidence: Mood, Individual Differences, Developmental and Social Aspects},
  year         = {2005},
}