Invitation: Graduation Presentations Friday, June 28th
Posted: 25 June 2019 09:07 PM
Administrator
Rank
Total Posts:  6
Joined  2018-12-11

This Friday eight students will present their graduation research projects.
You are of course all invited to attend.

When and where
Friday June 28th, 09:30 - 16:00
Gravensteen Building (Pieterskerkhof 6, Leiden), room 0.11

Schedule
09:30 - 10:05   David Lopez-Cotarelo
10:10 - 10:45   Giulio Barbero
10:50 - 11:25   Donna Schreuter
11:30 - 12:05   Christina Mason
12:05 - 12:45   Break
12:45 - 13:20   Mark de Reijer
13:25 - 14:00   Jichen Wu
14:05 - 14:40   Winke Wiegersma
14:45 - 15:20   Mark van Koningsveld  

Additional information
09:30 - 10:05
Student: David Lopez-Cotarelo
Title: Psychological attentional restoration through the haptic experience of water’s motions
Abstract: Living in a fast paced world, where productivity and accomplishment are commonly strived for is causing increased strain on the human mind. As evidenced by recent increases in the number of psychological disorders and decreases in perceived well-being of urban populations. The perception and experience of environments and their features can highly influence the states of mind in humans. For this reason, new approaches need to be taken to improve or redesign the experience of urban environments, as these are often inhabited by productive yet mentally fatigued populations. Experiencing features of natural environments has been demonstrated to induce positive emotions and reduced mental strain, which strongly link to perceived well-being. This study aims to make a call for innovative interventions that accessibly integrate and facilitate the exposure of natural features within urban environments. As such, this study investigates the potential for haptic experience of the material motion of water in restoring or improving psychological attention. To research this, participants had to complete a 3 minute long Stroop task that would require prolonged use of their directed attention. They would subsequently experience the motions of water by lying on a floating platform while being deprived of sight and hearing for 3 minutes. As alternative condition they would experience a control stimulus, lying on firm ground.  A Necker cube task was used to objectively assess their level of attention at different intervals, once after induction of attentional fatigue, and once again following exposure to conditioned stimulus. In addition, an EEG sensor was used to measure the level of attention over time during each interval that participants carried out the Necker cube task. Participants filled in a questionnaire providing an indication on their subjective state of restoration following their exposure to the stimulus from each condition. Upon ending the experiment they filled in a contextual questionnaire aiming to find out participants’ affinity with water-based activities and their engagement in any attentional demanding task preceding the experiment. The results showed no statistically significant changes in both Necker cube task scores and EEG sensed levels of attention, as such there is no conclusive evidence in support of an attentional restoration following the haptic experience of water’s motions. Nevertheless, the mean perceived restorative state ratings from the questionnaires did show a statistical significance, this shows that participants perceived a subjective restoration, but it cannot be attributed as attentional. Further studies should increase the amount of time in the induction of attentional fatigue to allow for a more visible and quantifiable effect of restoration.

10:10 - 10:45
Student: Giulio Barbero
Title: Intentionality and incorporation: understanding complex social interactions in a videogame
Abstract: Recursive mindreading is the human ability to understand and interpret nested mental states and beliefs held by one or multiple agents. While theory of mind represents a second level of intentionality (e.g.: “I believe that she believes something”), we can extend this chain to more complex relationships between several individual (e.g.: “I believe that she believes that he believe something” as an example of third level intentionality) and act accordingly. One of the main issues in the study of recursive intentional states is the difficulty to test human understanding. Current methodology involves several forms of storytelling and questionnaires. In all these cases the participants act as passive observers or listeners, never actively being involved into the actual social context they need to understand. In this research we want to study whether participants incorporated in the social environment, reconstructed in a video game, perform better than those who passively assist to a video presenting the same story.

1/3

Profile
 
 
Posted: 25 June 2019 09:11 PM   [ # 1 ]
Administrator
Rank
Total Posts:  6
Joined  2018-12-11

10:50 - 11:25
Student: Donna Schreuter
Title: Conformity with Conversational Assistants
Abstract: AI voice assistants are becoming increasingly popular in consumers daily lives. They are designed and developed to be more humanlike in their speech. This study investigates the effect this can have on ones conformity with an AI assistant. Solomon Asch’s series of psychological experiments in the 1950s demonstrated the power of conformity in social groups, and were later replicated with embodied robots. In light of our increasing reliance on AI assistants, this study investigates to what extent an individual will conform to a disembodied assistant, and if there is a difference between a group that interacted with a robotic voice and one that interacted with a humanlike voice. Participants completed a general knowledge quiz with the help of an AI assistant that communicated in text, a robotic voice or a humanlike voice. The assistant would attempt to subtly influence the individual’s final responses. We measured how often participants changed their answer to conform with the assistant. Results show a significant difference in conformity between the three groups. Participants conformed significantly more often to the assistant with a human voice than the one that communicated through text. While there is no significant difference between the two voice assistants, the results show that people are more likely to conform with an artificially intelligent assistant when it has a voice.

11:30 - 12:05
Student: Christina Mason
Title: Using Identity and Uncertainty to Model Opinion Dynamics in Online Social Networks
Abstract: The effects of social identity and certainty on social influence were studied using three agent based simulations (ABSs).  This was done to address how social influence affects attitudes in an online social media context, where social identity is implicit in the exchange of information.  It is hypothesized that the influence of other agents based on ingroup/outgroup perceptions can facilitate extremism and polarization under conditions of uncertainty.  The first two simulations isolated social influence and certainty as variables to see what effects these have on attitude formation.  Specifically, they were designed to address the fallbacks of previous models of opinion dynamics. These fallbacks were addressed to some extent, but the problems were not fully resolved.  The third combined the two to see if the limitations of both designs would be ameliorated with added complexity.  The combination proved to be moderating, and while stable opinion clusters form, extremism and polarization do not develop in the system without added forces. 

12:05 - 12:45
Break

12:45 - 13:20
Student: Mark de Reijer
Title: Calming pressure: a study through design on stress reduction by deep pressure touch contractions through wearable technology
Abstract: Stress (and anxiety) complaints tend to become more pervasive in today’s life [3, 21]. Reducing levels of stress is becoming of more importance for mental wellbeing due to its negative consequences for our health [7, 5]. Therefore, a way to lower our levels of stress would be invaluable. The crossover between stress and ever more present wearable technology show such devices would be promising. Deep Pressure Touch, a technique to reduce stress by adding pressure on the body, is known for its therapeutic calming effect [7, 8, 10, 25]. In this study, the researcher developed a wearable device, Premo, and tested its effectiveness in reducing stress levels while performing tasks. A tailored system had to be designed because of the lack of existing adaptive Deep Pressure Touch systems and the novel inclusion of the use of rhythm instead of continuous pressure.
This paper starts by outlining stress, deep pressure touch, related wearable technologies, rhythm and social synchronicity. Followed by highlighting the gap in the field leading up the design guidelines for such a device. In order to validate the hypotheses, a device was designed following the design guidelines which was tested on participants when performing a stressful task. The research sheds light on the design processes, potential strengths and advice for future studies.
In this paper it is presented how a research-through-design method contributes to the exploration of design for our mental health. It is also argued that the integration of psychological insights with wearable technology makes way for an engaging area of research and enables more meaningful technologies.

13:25 - 14:00
Student: Jichen Wu
Title: Art creation with a cross-domain mapping network
Abstract: Making relations between objects/concepts is believed to be crucial to human creativity. Relations are made by connecting relevant features of different objects/concepts. Different ways of connecting features will result in different related objects/concepts. I propose to use a neural network to map extracted features and map instances from one domain to new instances in a different domain. However, infinitely many mapping functions can be found without further limitations. to overcome this problem, a partially joint distribution is learned from marginal distributions of two separate domains. That is, two types of features are extracted from an instance: joint features in continuous space that are assumed to be shared between domains and separate features in discrete space that are connected differently based on different situations. The network is tested with two tasks: image-to-image mapping and audio-to-video mapping.

2/3

Profile
 
 
Posted: 25 June 2019 09:12 PM   [ # 2 ]
Administrator
Rank
Total Posts:  6
Joined  2018-12-11

14:05 - 14:40
Student: Winke Wiegersma
Title: Who-am-I-? An explorative investigation of a curated research method
Abstract: I can come up with eighty-one (and counting) possible answers to who I am, restricting myself to singular nouns. Which one is the ‘correct’ answer? Is there a ‘correct’ answer? This could have been a study into the multiplicity of possible answers. It could have been a quest to a single answer. It is neither. Rather, this is a study about the question. Why do I ask myself this question? How can I understand this question? What implications precede this question?
I constructed a who-am-I-? object; it is a four-dimensional cube. It is a complex matter; I tried to fit it in a neat shape, but it churns, twitches and squirms. How can I dissect such a strange research object? I decided to build a methodology. Appropriating from existing research methods, theories and styles, I curated a research methodology to investigate this question. Armed with this hybrid approach, I try to dissect each axis of the cube; the who, the am, the I and the ?.
Who-am-I-? is three things:
An investigation of a curated research method.
An explorative dissection of a question as a research object.
A self-experimentation of a researcher.

14:45 - 15:20
Student: Mark van Koningsveld
Title: The influence of Passive Haptic Learning on response times and error rates in a visual CRT task
Abstract: We have introduced Passive Haptic learning in a standard Choice Reaction time task that consisted of a pre-test, a distraction game and a post test. We looked both at the decrease changes in reaction time and improvement of error rates. Passive Haptic Learning
(PHL) is the concept of learning motor skills by receiving haptic information while performing a distractive task that asks for full attention. We created a glove-based and system that was able to provides haptic feedback corresponding to visual cues. After a pre- test where a baseline was established, both the PHL group and the control group played a game of snake. The PHL group received visual cues paired with vibrations in the corresponding
fingers. We found no significant improvement in mean reaction times in the PHL group when compared to the control group, although they did perform slightly better. We did however see that the error rates of the PHL group cut in half while those of the control group stayed almost the same. Although still not significant it is meaningful to do further research into the effect of PHL on motor control programs and uses outside sequential motor learning.

3/3

Profile