Skip to main content
Login | Suomeksi | På svenska | In English

Browsing by master's degree program "Cognitive Science"

Sort by: Order: Results:

  • Suoranta, Sanna (2022)
    This master's thesis investigates how the saliency of user interface components affect users when they choose between two options. Many services have moved online and they usually ask their users to accept web cookies that are either necessary for the functioning of the service or improve it. Users are enticed to operate to the advantage of the service provider by offering an attention-grabbing salient option for acceptance. For this thesis, software was developed, in which users were presented with three types of dialogues, with a difference in visibility between the two UI components. The user was offered two normal buttons to choose from, a normal button with a background-free button, or a normal button with a link. The experiment had 20 participants who were monitored using eye-tracking camera glasses, and the speed and accuracy of their tasks were measured using timestamps at the clicking of a button. The tasks were so easy that almost all participants were able to choose requested action even though the task was asked to be done as fast as possible. The layout of the buttons had no statistically significant effect on speed, but the text of the button had: the quickest was to select the button that contained the text OK. In the experiment, the marking of the areas of the direction of the gaze was not precise enough to determine where the user first looks. However, statistically significant the fastest was to complete the task where the most salient option was the correct option, regardless of whether the user looked at only this side or both sides of the dialogue, or those cases where the right answer was a positive option on the right side of the dialog. Surprisingly, without the ecologically valid framing created by the task, the mere visibility of the component was not a significant factor, but the text of the UI component was.
  • Lohilahti, Jonne Antti Kristian (2022)
    Tavoitteet. Tämän tutkimuksen tavoitteena on arvioida tunteiden havaitsemisen mahdollisuutta arkielämässä puettavien laitteiden ja koneoppimismallien avulla. Tunnetiloilla on tärkeä rooli päätöksenteossa, havaitsemisessa ja käyttäytymisessä, mikä tekee objektiivisesta tunnetilojen havaitsemisesta arvokkaan tavoitteen, sekä mahdollisten sovellusten että tunnetiloja koskevan ymmärryksen syventämisen kannalta. Tunnetiloihin usein liittyy mitattavissa olevia fysiologisia ja käyttäymisen muutoksia, mikä mahdollistaa koneoppimismallien kouluttamisen muutoksia aiheuttaneen tunnetilan havaitsemiseksi. Suurin osa tunteiden havaitsemiseen liittyvästä tutkimuksesta on toteutettu laboratorio-olosuhteissa käyttämällä tunteita herättäviä ärsykkeitä tai tehtäviä, mikä herättää kysymyksen siitä että yleistyvätkö näissä olosuhteissa saadut tulokset arkielämään. Vaikka puettavien laitteiden ja kännykkäkyselyiden kehittyminen on helpottanut aiheen tutkimista arkielämässä, tutkimusta tässä ympäristössä on vielä niukasti. Tässä tutkimuksessa itseraportoituja tunnetiloja ennustetaan koneoppimismallien avulla arkielämässä havaittavissa olevien tunnetilojen selvittämiseksi. Lisäksi tutkimuksessa käytetään mallintulkintamenetelmiä mallien hyödyntämien yhteyksien tunnistamiseksi. Metodit. Aineisto tätä tutkielmaa varten on peräisin tutkimuksesta joka suoritettiin osana Helsingin Yliopiston ja VTT:n Sisu at Work projektia, missä 82:ta tietotyöläistä neljästä suomalaisesta organisaatiosta tutkittiin kolmen viikon ajan. Osallistujilla oli jakson aikana käytettävissään mittalaitteet jotka mittasivat fotoplethysmografiaa (PPG), ihon sähkönjohtavuutta (EDA) ja kiihtyvyysanturi (ACC) signaaleita, lisäksi heille esitettiin kysymyksiä koetuista tunnetiloista kolmesti päivässä puhelinsovelluksen avulla. Signaalinkäsittelymenetelmiä sovellettiin signaaleissa esiintyvien liikeartefaktien ja muiden ongelmien korjaamiseksi. Sykettä (HR) ja sykevälinvaihtelua (HRV) kuvaavia piirteitä irroitettiin PPG signaalista, fysiologista aktivaatiota kuvaavia piirteitä EDA signaalista, sekä liikettä kuvaavia piirteitä ACC signaalista. Seuraavaksi koneoppimismalleja koulutettiin ennustamaan raportoituja tunnetiloja irroitetujen piirteiden avulla. Mallien suoriutumista vertailtiin suhteessa odotusarvoihin havaittavissa olevien tunnetilojen määrittämiseksi. Lisäksi permutaatiotärkeyttä sekä Shapley additive explanations (SHAP) arvoja hyödynnettiin malleille tärkeiden yhteyksien selvittämiseksi. Tulokset ja johtopäätökset. Mallit tunnetiloille virkeä, keskittynyt ja innostunut paransivat suoriutumistaan yli odotusarvon, joista mallit tunnetilalle virkeä paransivat suoriutumista tilastollisesti merkitsevästi. Permutaatiotärkeys korosti liike- ja HRV-piirteiden merkitystä, kun SHAP arvojen tarkastelu nosti esiin matalan liikkeen, matalan EDA:n, sekä korkean HRV:n merkityksen mallien ennusteille. Nämä tulokset ovat lupaavia korkean aktivaation positiivisten tunnetilojen havaitsemiselle arkielämässä, sekä nostavat esiin mahdollisia yhteyksiä jatkotutkimusta varten.
  • Ylätalo, Hanna (2022)
    Objective. Empathy is essential for successful collaboration. Empathic mechanisms partly rely on receiving sensory socioemotional information during social interactions, such as facial expressions. Today, computer-mediated communication (CMC) covers a large part of daily social environments. However, socioemotional information during CMC is restricted, which directly impacts social processes and therefore, different empathic skills may become beneficial in CMC environments compared to face-to-face interactions. The impacts of CMC on social processes are insufficiently understood and studies provide mixed results. Physiological synchrony is a useful tool to study underlying aspects of social interactions. In psychophysiology, physiological states can be evaluated according to physiological responses, such as changes in electrodermal activity (EDA), which is a measure of sympathic nervous system activity. EDA synchrony is connected to empathy and collaboration in several studies. The purpose of this study is to reveal connections between empathic skills, collaborative task performance and EDA synchrony in CMC environment. Methods. EDA signals of twenty-nine pairs were recorded during collaborative task performance in VR environment. Participants were unfamiliar with each other and could not see each other during performance. Before the experiment, they conducted two empathy tests: Interpersonal Reactivity Index (IRI) and Reading the Mind in the Eyes (RME). The performance was measured and connected with empathic abilities using statistical methods. EDA synchrony indices were calculated for each pair, and were statistically connected with empathy and task performance. Results and Conclusions. The results surprisingly showed that IRI subscale ’personal distress’ predicts collaborative task performance in VR environment. Personal distress reflects emotional sensitivity and is conneceted to social avoidance and maladaptive emotion regulation strategies. This result indicates that different social skills become beneficial in CMC environment, where participants cannot see each other, as in face- to-face collaboration. In addition, RME, which reflect skills in complex emotion recognition, was connected to performance on a trend level, which is supported by previous findings. EDA synchrony occured, but was not connected with either empathic skills or collaboration.
  • Ketola, Katriina (2021)
    Tarkastelen abrahamilaisen institutionaalisen jumalauskon kehittymistä, omaksumista ja säilymistä selittäviä kognitiivisia ja evolutiivisia mekanismeja 2010-luvun psykologisen tutkimuksen ja kognitiivisen uskontotieteen filosofien näkökulmista. Jumalauskon kehittymisen ovat mahdollistaneet seuraavat kognitiiviset tekijät: intuitiivinen ajattelutyyli, kognitiiviset vinoumat eli ontologiset kategoriasekaannukset, teleologinen ajattelu ja dualismi sekä skitsotypia. Nykyisen muotonsa jumalausko on kuitenkin saanut tuhansia vuosia kestäneen kulttuurillisen tiedon omaksumisen, säilyttämisen, muokkaamisen ja edelleen siirtämisen prosessien kautta. Yksilö omaksuu uskon jumalaan kulttuuriympäristöstään. Lisäksi yksilöllistä vaihtelua selittävät yllä eritellyt kognitiiviset tekijät. Yksilötasolla jumalauskon säilymistä motivoi eksistentiaalisen toimintakyvyn ylläpitäminen: usko toimii turvan lähteenä, maailmaa jäsentävänä mallina sekä mahdollisesti torjuu kuolemanpelkoa. Uskomiseen motivoitunut henkilö hakeutuu tilanteisiin, joissa hän kokee jumalan läsnäolon. Nämä kokemukset jumalan läsnäolosta vahvistavat uskoa entisestään. Yhteisötasolla jumalauskoa ylläpitää kyseenalaistamista vaikeuttava sosiaalinen paine, joka ilmenee epävakaissa elinolosuhteissa
  • Elonen, Jasmin (2021)
    To make fast and efficient decisions in changing environments, humans must plan their actions throughout their lives by maintaining and updating relevant information. Such goal-directed situations demand flexible adjustment of behavior and the suppression of task-irrelevant details. Executive dysfunctions in cognitive flexibility, working memory and inhibition have been related to aberrant prefrontal cortex functioning. The prefrontal cortex has previously been found to have an important role in these executive functions as a supervisory modulator and processor of information from posterior sensory brain areas. Structural and functional abnormalities in these brain areas have been found in attention deficit/hyperactivity disorder (ADHD). These impairments may be the underlying reason for problems in decision making and planning for people with ADHD. However how the brain’s oscillatory activity modulates different cognitive functions in areas relating to planning and decision making is still unclear. To investigate thisthe brain’s activity was measured with MEG while participants (21 ADHD patients, 28 controls) performed in Wisconsin card sorting test (WCST). WCST is a fast-paced task, where cards are sorted according to changing rule categories with the direction of feedback. The aim of this study was to investigate planning in adults with ADHD and a control group during WCST planning period and compare trials with previous feedback being correct or incorrect. Difference in the local neuronal activity in prefrontal and posterior areas were expected between the ADHD and control groups. Data-analysis and source modelling and reconstruction were conducted on the neuronal (MEG) data and structural (MRI) data. Statistical analyses were run for local neuronal amplitude dynamics and visualized within and between groups. The behavioral results of reaction times and hit rates did not show significant differences between groups. Clinical questionnaire scores did not correlate with reaction times. However, contrast of planning in correct-incorrect feedback trials within groups showed increased and decreased brain activity in delta, theta, alpha and beta oscillations. The control group showed activity in frontal, parietal, temporal, and occipital regions related to default mode, somatomotor with increased activity in the beta and alpha bands. For the ADHD group greatest positive activity was seen in beta band in frontal and parietal areas, but also in occipital regions. These activation sites were mostly related to dorsal and ventral attention and somatomotor networks but also to default mode and visual networks. Mostly temporal activity of suppressed delta, theta and alpha oscillations in the lateral areas was seen in the between groups comparison. These areas related to ventral attention and somatomotor networks. Impaired neuronal activity in the ADHD group was seen in weaker increased beta than the control group and the missing increased alpha oscillations. It remains for the future studies to interpret the roles of this oscillatory activity but direction towards impairments in cognitive functions like flexibility, working memory and inhibition in planning in ADHD. These data also suggest that planning in the WCST needs the flexible modulation of many cognitive functions and processes that are modulated by increased alpha and beta oscillations and the suppressed delta and theta oscillations.
  • Nurmi, Miska Juhani (2021)
    Objectives The purpose of this thesis is to consider what the cognitive models of online causal learning are and what they have to offer for the interactive AI approach. In this thesis, an interactive AI system is considered one that focuses on understanding and collaborating with a human user and which can therefore benefit from cognitive models. The general overview of the models is given by replicating some of the computational results of Bramley et al. (2017) which explored cognitive models for online causal learning. The earlier paper contained four models on how people might learn their causal beliefs, and five models on how people might choose where they place their tests, also known as interventions. Thesis also discusses the implications that the replicated models have for interactive AI, both by considering how these models could be better extended into the interactive AI framework, but also by considering a simple AI based system that could make use of such models. Replication The replication was done by reimplementing the original models of Bramley et al. in R and by reproducing the corresponding figures. Out of the four models used for causal belief updating, two were successfully replicated so that the results corresponded to the original paper. It is not certain why the two other models could not be replicated, and the task is left open for future work. Out of the five intervention choice models, four were implemented and three successfully replicated. One of the models was very close to the original results, but this thesis could not conclude whether it fully reproduces the original results. Implications The simple AI model proposed in this performed poorly but was able to show that in theory, an interactive AI system that incorporates such a model might be feasible in the future with further development. Some recommendations to better extend the replicated models into the interactive AI framework were made. Main recommendations were that a better model on how people might choose where they focus their local attention is needed. Furthermore, it should be ensured that the models approximate human behaviour in larger graphs as well.
  • Stenius, Thomas Christian (2021)
    There is a persistent belief among design scholars that design and science are fundamentally different activities. While there are historical reasons for this, the academic literature surrounding both design and scientific cognition does not fully support this notion. Against this background, the aim of this thesis is to examine how designers think while designing and the ways in which design thinking may resemble scientific thinking. For this study, verbal protocol analysis was chosen as the method. Five experienced architects were asked to perform a design task for approximately an hour, and think aloud concurrently while designing. Their verbal output was then transcribed and coded inductively, and analyzed using both the video and verbal data in parallel without directly utilizing any existing coding scheme or framework. Subjects used basic sketching tools during the sessions, which were conducted remotely over video conferencing software due to the ongoing COVID-19 pandemic. After multiple iterations of inductively coding the data, behavioral patterns consisting of three main activities were identified: framing (mentally defining the boundaries of a specific topic of focus), creation (generating and implementing ideas), and evaluation (evaluating previous designs). These patterns, or design activity sequences, formed the basis for the design activity model (DAM) that was developed and used in this study. Five different sequences were identified: full loops, extensions, partial loops, reverse loops, and parallel sequences. Subjects tended to start and end any design action with framing, which is why these sequences are called loops. They varied in duration but were generally short, ranging from a few seconds to a minute or two. On average, subjects would perform roughly two sequences per minute. These, however, tended not to be evenly distributed over the entire duration of the design task. The cognitive activities of the subjects were also contrasted with cognitive activities found in science, including deductive, inductive, and abductive reasoning, analogies, categorization, causal reasoning, distributed reasoning, and hypothetico-deductive reasoning. Many similarities were identified, but not all of them could be observed directly from the data and had to be inferred from the context, as subjects did not always verbally express what their exact reasoning patterns were. The empirical analysis suggests that designers utilize specific sequences of actions while designing. The study suggests, albeit not conclusively, that there are similarities between design and scientific thinking, which is in contrast to the popular belief among design scholars that they are fundamentally different. To complement this study, future avenues of research are suggested. An additional contribution of this study is the DAM itself and its coding scheme.
  • Makkonen, Emilia (2021)
    Transcranial direct current stimulation (tDCS) is a method that in recent years has gained the attention of researchers. In certain contexts, tDCS can be utilised in modulating brain function and cognitive performance, and it has been found to modulate symptoms and neural correlates of some pathologies. Due to differences in protocols between studies, the literature is inconclusive on the effects of tDCS on some brain areas and networks. In this study the effects of tDCS directed to medial prefrontal cortex (mPFC) on functional networks of the brain in resting state are examined. 36 healthy female subjects were divided into control and tDCS-groups of equal sizes. Functional networks were examined using functional magnetic resonance imaging (fMRI) in resting state before one stimulation session of 20 minutes with 2 mA current or sham stimulation, and after the session. Functional networks were identified with independent component analysis (ICA), which was performed on the set of images with both groups combined. The groups were then compared on a network level. The tDCS group showed increased functional connectivity in and around the left anterior cingulate cortex (ACC) in language network compared to the sham group. Other networks identified by ICA that showed no statistically significant differences between groups were posterior and anterior default mode, dorsal attention, sensorimotor, orbitofrontal, left and right frontoparietal, medial and lateral visual network, basal ganglia and the cerebellum. The results of this study were partially in line with previous studies that have found increases in the functional connectivity of brain regions associated with reward processing. There is a need for systematic examination of the effects of different tDCS protocols on functional networks in future studies.
  • Halmetoja, Saara (2021)
    Kaksikielisyyden yhdentymishypoteesi (engl. convergence hypothesis) esittää, että kehittyvän kielitaidon myötä ensimmäisen ja toisen kielen käsittelyyn käytetään jaettuja aivoalueita. On avoin kysymys, miten kaksikieliset käsittelevät sanatason kieliopillista eli morfosyntaktista tietoa yksikielisiin verrattuna. On esitetty, että äidinkielen morfosyntaktista ja sanastollista tietoa käsitellään aivotasolla eri mekanismein ja osittain eri aivoalueilla. Morfosyntaksin käsittelyn aivoperustaa on tutkittu paljon säännöllisten ja epäsäännöllisten verbien avulla, mutta aiheen tutkimus muilla sanaluokilla on jäänyt vähemmälle huomiolle. Taivutetut substantiivit sisältävät sekä syntaktista että sanastollista tietoa. Tässä pro gradu -tutkimuksessa selvitettiin ensimmäisen ja toisen kielen taivutettujen sanojen käsittelyä aivoissa tapahtumasidonnaisten jännitevasteiden avulla. Äidinkieleltään englanninkielisiltä yksikielisiltä (n=20) ja kaksikielisiltä (n=17), jotka olivat äidinkieleltään suomenkielisiä ja englantia toisena kielenä puhuvia, mitattiin aivosähkökäyrää passiivisen kuuntelun tilanteessa. Koehenkilöille esitettiin puhuttuja englannin- ja suomenkielisiä sanoja ja akustisesti samankaltaisia pseudosanoja, joista osa oli taivutettu suomen ja englannin monikon päättein (/s/ ja /t/). Morfosyntaksin käsittelyä eri kielitaustoilla tutkittiin vertailemalla taivutuspäätteille syntyneitä jännitevasteita. Englannin monikkopäätteelle /s/ syntyneet vasteet olivat kielestä riippumatta voimakkaampia, kun sanavartalona oli oikean sanan sijaan merkityksetön pseudosana. Tämä vaikutus jatkui pidempään, kun kyseessä oli oikein taivutettu englanninkielinen sana. Tulos on samansuuntainen kuin aiemmissa yksikielisillä tehdyissä tutkimuksissa. Vasteet suomen kielen päätteelle /t/ eivät tuottaneet vastaavaa eroa vasteissa, vaan suomea heikosti taitavilla englanninkielisillä aikainen vaste oikein taivutetuille suomen sanoille oli suurempi kuin taivutuspäätteellisille pseudosanoille. Sama vaikutus esiintyi myöhäisessä vasteessa kielitaustasta riippumatta. Äidinkielelle syntyneet aivovasteet eivät siis olleet täysin yhteneviä kaksikielisten ja yksikielisten välillä. Yksikielisillä havaittiin englanninkielisten sanojen yhteydessä vastakkainen vaikutus, jossa aikainen vaste oli suurempi, kun päätteeseen liitetyllä sanalla ei ollut merkitystä. Kokonaisuudessaan tulokset viittaavat siihen, että sujuvataitoiset vieraan kielen puhujat käsittelevät morfosyntaksia yhtenevästi syntyperäisten puhujien kanssa, mikä tukee yhdentymishypoteesia.
  • Lehtonen, Noora (2022)
    Friendship can be defined as long-term social bonds between non-kin, characterized by mutual affection and support. Friendship is often studied in dyadic interactions or in ego-networks, but less is known about the group level processes of friendship. This thesis investigates factors related to maintenance of adult friendships over longer periods of time, using real-life, retrospective data of friendship groups formed during the study years and maintained well into adulthood. Based on group research in cognitive and social psychology and evolutionary theories on friendship and cooperation, I am especially interested in the effect of group size and group bonding on group success, as well as the role of possible gender differences. Using data from the Fraternity Friendship Study (N = 284), collected from fraternity alumni members, I investigate factors that help adult friendship groups succeed over time. The friendship groups were between 5 and 18 members in size and were formed 12-24 years ago. The survey data regard aspects of fraternity activities and friendship groups both at the time of studies and at present. Group success is measured as group bonding (the Inclusion-of-Other-in-Self or IOS scale), reflecting the members’ emotional closeness towards the group, and meeting frequency. Qualitative aspects of group functioning are also examined. Results show that group success was related to group size at the time of group formation, as well as with qualitative aspects of the group, such as information flow, perceived homophily, and perceived attractiveness of the friends in the group. Group bonding moderated the effect of group size on meeting frequency, so that smaller groups benefitted more of bonding. Group bonding was also itself an important factor for how active the groups remained over the years. Group success did not differ between male and female friendship groups: bonding, meeting frequency, and group size did not vary by gender. However, group size had a slightly larger effect for the maintenance of female friendships, compared to male groups: especially small female groups were more likely to lose contact. Overall, a group size of approximately 10 members or more, roughly corresponding to sympathy group in the layered social network model, was found to be beneficial for group success.