Pregled bibliografske jedinice broj: 868779
An image is worth a thousand words, but what of numbers? The impact of multi-modal processing on response times and judgements of confidence in base-rate tasks
An image is worth a thousand words, but what of numbers? The impact of multi-modal processing on response times and judgements of confidence in base-rate tasks // XXIII Naučni skup Empirijska istraživanja u psihologiji
Beograd: Laboratorija za eksperimentalnu psihologiju Beograd, 2017. str. 67-68 (predavanje, međunarodna recenzija, sažetak, znanstveni)
CROSBI ID: 868779 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
An image is worth a thousand words, but what of numbers? The impact of multi-modal processing on response times and judgements of confidence in base-rate tasks
Autori
Dujmović, Marin ; Valerjev, Pavle
Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni
Izvornik
XXIII Naučni skup Empirijska istraživanja u psihologiji
/ - Beograd : Laboratorija za eksperimentalnu psihologiju Beograd, 2017, 67-68
ISBN
978-86-6427-048-9
Skup
XXIII Naučni skup Empirijska istraživanja u psihologiji
Mjesto i datum
Beograd, Srbija, 24.03.2017. - 26.03.2017
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
base rate, metacognition, confidence judgment, dual processing, multi-modal processing
Sažetak
This study continues a string of research into the relationship between conflict detection and metacognitive judgments. For high conflict tasks we observed longer response times and lower levels of confidence. Our current research introduces images into the base-rate task. We used a simple base-rate task, e.g.: “Person A is tall”, followed by group information “The group consists of 850 basketball players, and 150 physicians”. Participants have to choose the subgroup from which the person was most likely randomly chosen. The base rate can be consistent or inconsistent with the intuitive answer. In one set of situations the image represented a base rate consistent or inconsistent with the intuitive answer, and in the second set, the image was accompanied by the usual base rate number (multi-modal situation). In the second set of situations the image was an equivalent representation of the ratio, or it was designed to mitigate the base rate by representing a 60% decrease in the ratio. In both of these situations base rates and images were in conflict with intuitive responses. Consequently, four distinct levels of conflict were induced before participants (N=35) made their decisions. After each decision participants made a judgment of confidence ranging from 50% (guessing) to 100% (complete confidence). As expected, one-way ANOVA showed a significant effect of conflict level on response times (F(3, 102)=6.71, p<.01) with higher conflict prolonging response times. The same effect was not found for confidence ratings, probably due to extremely high judgments overall. Higher conflict levels lead to lower rates of intuitive responses (χ2(3)=19.15, p<.01). We also compared induced levels of conflict by images compared to numeric base rates using data from our recent study. Calculating a 2(image-base rate)x2(consistency) ANOVA we found a strong overall effect of consistency (F(1, 63)=30.79, p<.01) with slower responses for conflict items. At the same time, the group by consistency interaction was also significant (F(1, 63)=7.19, p<.01). Conflict items with images increased response times to a lesser degree than the usual numeric base-rate items. The same effects (consistency F(1, 63)=9.67, p<.01 ; group by consistency F(1, 63)=4.09, p<.05) were found for confidence ratings. Conflict items reduced confidence ratings significantly more for classic numeric base-rate items than for image based items.
Izvorni jezik
Engleski
Znanstvena područja
Psihologija
POVEZANOST RADA
Projekti:
Projekt 4139 Hrvatska zaklada za znanost
Ustanove:
Sveučilište u Zadru
Profili:
Pavle Valerjev
(autor)