Browsing by Subject "working memory"

Human perception is seriously limited in relation to the countless stimuli of one’s surroundings. It is guided by attention, which can be directed only at a fraction of all available stimuli at a given moment. The topic of this thesis is a phenomenon illustrating the constrained nature of attention called the attentional blink. One of the goals is to consider the origin and mechanisms behind the phenomenon as well as the current applications of 1) the phenomenon itself and 2) the methods used in attentional blink research. The attentional blink occurs in a rapid succession of stimuli, which unveils the limits of temporal capacity of attention. As a consequence of the blink, some of the target stimuli do not reach conscious processing. The phenomenon has sparked wide interest since the 1980’s, but the mechanisms, operating principle and especially the neural correlates of attentional blink are still under debate, which makes the topic relevant even today. Attentional blink is most commonly studied with the rapid serial visual presentation –method (RSVP). This method, however, can be applied to many more contexts as well, some of which are addressed here. The thesis is a review covering 32 articles and publications of the attentional blink or topics directly related to it. They include only reports of studies done on healthy individuals and reviews of such experiments. Although there is much more literature on the topic besides the ones chosen here, these 32 provide a critical, yet comprehensive review of an attentional limit called the attentional blink. Attentional blink is commonly known as a consistent and universally occurring phenomenon, during which attentional resources are temporarily suppressed. According to most theories it occurs due to choosing specific target items to be processed in working memory, though contradicting evidence exists. Somewhat surprisingly, there are individual differences in the duration and magnitude of the blink, which are affected by the processing speed of working memory and the ability to update its contents. Thus the blink does not appear exactly the same way for everyone. Actual applications of the blink seem to be very few in number, possibly due to limited benefits of application or the specificity of the phenomenon. The RSVP-method used in attentional blink research is more widely applied in modern technology, for example in text presentation. For any application to be useful, it needs to reduce the overload on attentional and working memory systems in order to heighten perception. However, to be able to utilize attentional blink and RSVP maximally, more research is needed.

The aim of the study. Working memory (WM) is a cognitive function that relies on the neurotransmitter dopamine. WM has multiple subfunctions: maintaining relevant information, ignoring distractors, and updating the information when needed. Information maintenance and ignoring irrelevant stimuli are associated with brain activity in the prefrontal cortex and updating information with the striatum. A connection between polymorphisms within the dopaminergic genes COMT (COMT Val158Met polymorphism) and DRD2/ANKK1 (Taq1A polymorphism), and WM performance has been established in previous studies, and these genes seem to impact dopamine signaling in the prefrontal cortex and striatum, respectively. The present study investigates associations between dopaminergic gene polymorphisms and WM performance. The main research questions include the main effects and interactions that the genes have to WM on performance overall and WM stability and flexibility in particular. The study investigates also whether obesity, diet, and age affect WM performance, as there are indications for these from previous studies. Methods. Data from three separate cross-sectional studies with a total of 244 participants was used. A computer-based visual working memory task was used to assess the WM performance measures accuracy and reaction time. Other measurements included blood measures, Body Mass Index (BMI), and the Dietary Fat and free Sugar Questionnaire (DFS). Results and conclusion. In WM accuracy analyses, there was a significant main effect for the task condition as well as interactions between the COMT*task condition and COMT*Taq1A*task condition. In the three-way interaction analysis for WM accuracy, the most beneficial combination was Met/Met & A1- in the task condition measuring the stability, and Val/Met & A1+ in the task condition measuring the flexibility. This result gives support for established knowledge about how Met/Met combined with A1- is beneficial especially for WM stability and maintenance. Also, the inverted-U shape theory and previous studies' results of A1+ being beneficial in executive functions updating tasks, give support to the result regarding WM flexibility. In the three-way interaction reaction time analysis, the combination of Val/Met & A1+ was the fastest in every task condition. However, the significant differences were mainly between A1- and A1+ genotypes when combined with Val/Met and between Met/Met and Val/Met genotypes when combined with A1-. The two-way interaction between COMT Val158Met and task condition did not remain significant in pairwise comparisons. In reaction time analyses, COMT Val158Met and Taq1A had a significant main effect where reaction times followed the order: Met/Met < Val/Val < Val/Met, with Val/Met being significantly slower than the others and in Taq1A A1+ < A1-, with a significant difference. Adding age, BMI, and DFS to the analyses did not affect the significant/non- significant main effects or interactions in the analyses. The results of this thesis converge with the previous knowledge about these genotypes having an interacting effect on working memory stability and flexibility. However, these effects are complex to interpret as the results and their directions differed between the task conditions and outcome measurements. In the main outcome variable (accuracy), the result regarding the WM stability is well in line with previous literature about the Met/Met & A1- combination being beneficial to WM performance, and the result about Val/Met & A1+ being beneficial for WM flexibility gets preliminary support from executive function studies, but also opens new research avenues regarding the WM flexibility subfunction.

Tavoitteet. Tässä tutkimuksessa oli tavoitteena tutkia, miten 5- ja 9- vuotiaiden lasten lyhytkestoisen muistin toimintaa mittaavat testien tulokset ovat yhteydessä aikuisena 40- vuotiaiden työmuistisuoriutumiseen sekä visuaaliseen ja verbaaliseen päättelykykyyn. Tutkimuskysymykset olivat: 1) Miten lapsuuden auditiivisen ja visuaalisen sarjamuistin tehtävissä suoriutuminen on yhteydessä aikuisuuden auditiivisverbaaliseen työmuistisuoriutumiseen? 2) Miten lapsuuden auditiivisen ja visuaalisen sarjamuistin tehtävissä suoriutuminen on yhteydessä työmuistisuoriutumiseen, jossa informaatiota manipuloidaan? 3) Miten lapsuuden auditiivisen ja visuaalisen sarjamuistin tehtävissä suoriutuminen on yhteydessä aikuisuuden kielelliseen ja visuaaliseen päättelykykyyn? Menetelmät. Tutkimusaineistona oli Kognition pitkäaikaistutkimus (KOPUTUS), jossa oli vuosina 1971- 1974 syntyneitä syntymäriskiryhmiin kuuluvia lapsia sekä kontrolliryhmä. Tähän tutkimukseen otettiin mukaan 5-, 9-, ja 40- vuotiaiden tutkimustulokset (N=301). Lapsuudessa tutkittavat suorittivat lasten kommunikaatiota ja kielellisiä kykyjä mittaavan The Illinois Test of Psycholinguistic Abilities (ITPA)- testin. Muuttujiksi valittiin lyhytkestoista muistia mittaavat testit: visuaalinen sarjamuisti ja auditiivinen sarjamuisti. Aikuisuudessa tutkittavat suorittivat WAIS- IV:n, josta muuttujiksi valittiin työmuistia mittaava perusosatehtävä Numerosarjat sekä kielellisen ja visuaalisen päättelyn indeksit (VCI ja PRI). Taustamuuttujiksi tutkimukseen otettiin tutkittavien oma korkein koulutus, äidin korkein koulutus ja turvapisteet 9- vuotiaana. Tutkimustulokset analysoitiin rakenneyhtälömallien avulla. Tulokset ja johtopäätökset. Tulokset osoittivat, että 5- vuoden ikäisten ja 9- vuoden ikäisten auditiivisen lyhytkestoisen muistin suoriutumiset erikseen ennustivat aikuisuuden työmuistisuoriutumista. Lisäksi 9- vuoden ikäisten auditiivisella lyhytkestoisella muistilla voitiin ennustaa myös monimutkaisempaa työmuistin toimintaa. Visuaalisella lyhytkestoisella muistilla lapsuudessa taas voitiin ennustaa visuaalista päättelykykyä aikuisena. Tuloksista voidaan päätellä, että lasten lyhytkestoisen muistin tutkiminen on tärkeää, koska sillä voidaan ennustaa aikuisuuden työmuistin toimintaa ja visuaalisen tiedon käsittelyä sekä mahdollisia heikentymiä näissä varhain.

Objectives. The sentence span task is a complex working memory span task used for estimating total working memory capacity for both processing (sentence comprehension) and storage (remembering a set of words). Several traditional models of working memory suggest that performance on these tasks relies on phonological short-term storage. However, long-term memory effects as well as the effects of expertise and strategies have challenged this view. This study uses a working memory task that aids the creation of retrieval structures in the form of stories, which have been shown to form integrated structures in longterm memory. The research question is whether sentence and story contexts boost memory performance in a complex working memory task. The hypothesis is that storage of the words in the task takes place in long-term memory. Evidence of this would be better recall for words as parts of sentences than for separate words, and, particularly, a beneficial effect for words as part of an organized story. Methods. Twenty stories consisting of five sentences each were constructed, and the stimuli in all experimental conditions were based on these sentences and sentence-final words, reordered and recombined for the other conditions. Participants read aloud sets of five sentences that either formed a story or not. In one condition they had to report all the last words at the end of the set, in another, they memorised an additional separate word with each sentence. The sentences were presented on the screen one word at a time (500 ms). After the presentation of each sentence, the participant verified a statement about the sentence. After five sentences, the participant repeated back the words in correct positions. Experiment 1 (n=16) used immediate recall, experiment 2 (n=21) both immediate recall and recall after a distraction interval (the operation span task). In experiment 2 a distracting mental arithmetic task was presented instead of recall in half of the trials, and an individual word was added before each sentence in the two experimental conditions when the participants were to memorize the sentence final words. Subjects also performed a listening span task (in exp.1) or an operation span task (exp.2) to allow comparison of the estimated span and performance in the story task. Results were analysed using correlations, repeated measures ANOVA and a chi-square goodness of fit test on the distribution of errors. Results and discussion. Both the relatedness of the sentences (the story condition) and the inclusion of the words into sentences helped memory. An interaction showed that the story condition had a greater effect on last words than separate words. The beneficial effect of the story was shown in all serial positions. The effects remained in delayed recall. When the sentences formed stories, performance in verification of the statements about sentence context was better. This, as well as the differing distributions of errors in different experimental conditions, suggest different levels of representation are in use in the different conditions. In the story condition, the nature of these representations could be in the form of an organized memory structure, a situation model. The other working memory tasks had only few week correlations to the story task. This could indicate that different processes are in use in the tasks. The results do not support short-term phonological storage, but instead are compatible with the words being encoded to LTM during the task.