When and where?

The time and location for each public presentation on Monday June 29:

• Anne Schröder, 11:15h, online via this Zoom link.
• Emile Velthuis, 15:00h, Gorlaeus building room BE 0.09.
• Timothy Schuur, 16:00h, Gorlaeus building room BE 0.09.

The Gorlaeus building is located at Einsteinweg 55. It is the new white building that says "Faculty of Science" on top. The presentations are public, all are invited to attend.

Anne Schröder

Seeing Through a Feminist Lens: How Female Character Visualisation Shapes Player Interpretation in Video Games
Starts at: 11:15h
Zoom link: https://universiteitleiden.zoom.us/j/62373546420?pwd=YBVmXUagPAVgxH8uXcpndbmnpPIHxb.1

This study examines how different approaches to female character visualisation in video games shape the way players interpret, emotionally engage with, and make sense of those characters. While research has shown that traditional, hypersexualised designs contribute to female players feeling excluded from gaming spaces, less is known about the specific mechanisms through which visual design produces these effects at the level of player reception. Using a feminist data visualisation lens (D'Ignazio & Klein, 2016), this research addresses that gap through a triangulated, mixed-methods design. Twelve mixed-gender adult gamers responded to four existing game characters, two traditionally designed and two feminist-inspired, through vignette-based semi-structured interviews and a Likert-scale survey, analysed across three dimensions: interpretation, emotional engagement, and sense-making. Six themes were found. Feminist-inspired designs enabled coherent character reading, produced identification in female players, and were perceived as broadly inclusive. Traditional designs obstructed character reading, produced felt exclusion in female players, and were consistently read as addressed to a specific male audience. The study identifies two contributions the existing literature does not contain. First, the gender asymmetry in response is structural rather than a difference in sensitivity: many male players engage with characters as tools for gameplay rather than as figures of identification, a position that insulates them from the exclusionary signals female players detect. Second, feminist-inspired design operates through subtraction rather than addition, succeeding by removing the elements that compete with the character for the viewer's attention rather than by adding a progressive marker. Inclusive character design is therefore a lower and more actionable threshold than the literature often implies.

Thesis advisors: Maarten Lamers and Zane Kripe

Emile Velthuis

Calculating Individual Notes’ Contribution to the Consonance/Dissonance Perception of a Vertical Note Set
Starts at: 15:00h
Location: Gorlaeus building room BE 0.09

With our study we research approaches for creating note-specific consonance/dissonance models for analysing vertical sets of notes. We do this by proposing adaptations to the best-performing existing consonance/dissonance models for simultanious sounding notes. We show that the roughness models from Hutchinson & Knopoff, Sethares, and Vassilakis can be transformed using the general rule that the roughness caused by interacting partials can be equally divided among the notes to which the partials belong. We have fully adapted the Hutchinson & Knopoff model, including accounting for coincident partials. For harmonicity/periodicity, the root-based models from Parncutt and Milne are seen as unsuitable since their calculations rely on a strongest (virtual) root, which causes the proposed note-specific method to be unstable and exception-prone. The Harrison & Pearce model seemed unsuitable since it has a strong negative correlation with the cardinality of the chord. For the Stolzenburg model, a note-specific variant has been proposed that has a comparable performance to the original model and has a strong correlation with the note-specific variant of the Hutchinson & Knopoff model, although the correlation is less strong than the correlation of those original models. For the familiarity models, it is argued that the calculation is not decomposable other than that every pitch class has an equal contribution in defining a unique chord class. Furthermore, two other methods have been proposed that use the output of the models instead of transforming their calculation. However, since they compare subsets of the note set, the context of the measures changes, which can lead to distorted note-specific values.

Thesis advisors: Edwin van der Heide & Frieder Stolzenburg (Harz University of Applied Sciences)

Timothy Schuur

THE PROVOCATIVE COACH — Ego-Threatening versus Autonomy-Supportive Feedback in a Real-Time Running Coaching Application: A Pilot Study
Starts at: 16:00h
Location: Gorlaeus building room BE 0.09

Real-time digital coaching for endurance running has predominantly relied on supportive or neutral feedback styles. This pilot study examines a counterintuitive alternative: ego-threatening feedback, which directly challenges the athlete’s competence relative to their goal. Drawing on Psychological Reactance Theory and the Dual Process framework, the thesis proposes that such cues may trigger an automatic fight impulse thatreduces reliance on deliberate effort regulation, potentially improving pace under acute fatigue. General Self-Efficacy is theorised to moderate this effect.
A within-subjects, counterbalanced field study compared autonomy-supportive and ego-threatening auditory coaching cues across two 5-kilometre runs. Sixteen recreational runners were recruited; nine were excluded post-hoc due to insufficient pace challenge, structural pace deficits, or incomplete data, resulting in a sample of n= 7. Outcome variables included average running pace (GPS-derived), perceived cognitive processing, perceived psychological stress, and motivational reactance.
No hypothesis was statistically confirmed. Average pace did not differ significantly between conditions (MAS = 311.7 s/km vs. MET = 314.6 s/km, p=.203, r=.48), with five of seven participants running faster under the autonomy-supportive condition.
The cognitive processing scale failed under ego-threatening conditions (α=−3.525), leaving H2 unresolved. Perceived  psychological stress was directionally higher under ego-threatening feedback for all participants who showed any difference (r=
.70), though this did not reach statistical significance (p= .066). Qualitative data suggest ego-threatening cues may trigger a brief acceleration that fades before accumulating into a lower overall pace.
The primary contribution of this study is methodological. Testing a novel and psychologically complex intervention in an uncontrolled field setting before its basic mechanism has been established under controlled conditions proved overambitious. Future work should prioritise a treadmill-based protocol that eliminates GPS noise and environmental confounds, with designs capable of capturing cue-level behavioural responses, strict counterbalancing, and a sample with verified pace-goal experience.

Thesis advisors: Edwin van der Heide & Bart Roelands (Vrije Universiteit Brussel)