• Eva F, Eva M, evaf.app
• Resonance Analyzer
• UGA Innovation Gateway I-Corps Accelerator program

A decade-long research and design project building accessible, evidence-based tools for voice feminization and masculinization training for transgender people. At its peak, the platform served 17,000 active users worldwide across mobile and web.

Eva F, Eva M, evaf.app

The EVA app development began as a design problem: could a structured, self-guided, visually mediated learning system provide effective voice training for transgender people who could not access or afford one-on-one speech therapy? As co-founder and Chief Creative Officer of VoxPop, LLC, I led all visual identity, product design, and user experience for the resulting platform from its founding in 2013 through its dissolution a decade later.

Resonance Analyzer

The platform I designed operated as a layered learning system. Each Program — EvaF for voice feminization, EvaM for masculinization — contained Courses of ten to twenty Lessons, each structured around three instructional videos and real-time scored exercises providing immediate visual feedback on voice parameters including pitch, breath, and resonance. A companion web-based subscription platform, evaf.app, extended the offering with additional content, a community layer, and live interactive webinars. At their peak, the apps reached 17,000 active users in more than 34 countries, averaging approximately 40 downloads per day, with coverage in the Smithsonian, The Guardian, VICE, National Public Radio, and BuzzFeed, among others. Through mid-2022, I produced and edited all video content for both the mobile apps and the web platform; I then hired and trained a video contractor to take on that role.

Beginning in 2020, I directed the development of two new features. The first was a refinement to the existing Pitch Tracker tool: I designed the UI for a "Pin High/Low Frequencies" function that allowed users to mark their personal pitch range endpoints and track progress over time — a functionally specific addition that gave users a persistent visual record of their own trajectory. The second was a Resonance Analyzer prototype: a real-time formant frequency visualizer capable of indicating whether a voice reads as masculine or feminine, grounded in peer-reviewed voice science data and developed in collaboration with voice scientist Ron Scherer, PhD, Professor Emeritus at Bowling Green State University. I designed the UI and UX specification for both a standalone and an in-exercise version. The prototype was built but did not reach deployment before the company's closure.

UGA Innovation Gateway I-Corps Accelerator program

In Spring 2021, VoxPop was accepted into the UGA Innovation Gateway I-Corps Accelerator program. I served as Academic Lead alongside co-founder Kathe Perez as Entrepreneurial Lead. Over eight weeks of structured hypothesis testing, we conducted more than 100 interviews with users, speech-language pathologists, advocates, and conference communities, refining three distinct user archetypes: people newly beginning transition, those mid-process, and those living post-transition who continued to experience voice dysphoria. The analysis confirmed that the Serviceable Addressable Market of $1.5 billion was too small to attract venture capital at the scale of investment the product required. I presented the VoxPop investor pitch on March 22, 2021, at the UGA Innovation Gateway before an audience of entrepreneurs and potential investors, and three ongoing business and research mentorships followed. ## Small Business Innovation Research Grant

The central outcome of I-Corps coaching was a strategic reframe: from venture capital toward federal research funding, and from a commercial product toward a clinically validated, gamified biofeedback intervention for voice dysphoria. With support from UGA Innovation Gateway mentors, we assembled a research team and began drafting a Small Business Innovation Research grant. I completed required CITI Program training in social/behavioral and clinical research practice, and we submitted IRB applications to UGA (Phase I) and Sterling, a commercial IRB (Phase II). In summer 2021, we submitted a $2.25 million SBIR Fast-Track application to the NIH — PA-21-259 — under the title “Eva the Voice Ninja: the voice trainer in the palm of your hand.” The NIH rejected the initial submission on a classification technicality, ruling that Phase II constituted a clinical trial. We rewrote the proposal accordingly. A second submission was rejected in November 2022; revision based on reviewer feedback was underway in early 2023.

The third revision was never submitted. In early 2023, co-founder and co-PI Kathe Perez withdrew from the project following the Marshall Fire, a Colorado wildfire that had destroyed her home. The loss made continued collaboration untenable, and unresolved questions around intellectual property ownership made it legally impossible to proceed without her. We made the joint decision to terminate the research project, discontinue development, and dissolve VoxPop LLC. I spent the greater part of 2023 on the administrative work of closure: delisting EvaF and EvaM from the App Store and Google Play, shutting down evaf.app, closing SaaS, financial, and platform accounts across multiple services, and ensuring current users received appropriate notice and transition support. The company was dissolved at the end of 2023, ten years after its founding.

• Research Proposal
• Data Visualization Pilot Workshop

Interdisciplinary project combining genomics, art, and graphic design to develop data visualization techniques making complex scientific data accessible to non-scientists. Co-PI with Chung-Jui Tsai (Warnell / Institute of Bioinformatics, UGA). All four NSF peer reviewers praised the science-art integration and design education component. Revision in progress.

Research Proposal

The Scientific Question Aspen trees cover more of the earth's surface than almost any other tree species, yet they cannot interbreed with their closest relatives. Cottonwoods, balsam poplars, and aspens all belong to the genus _Populus_ and share strikingly similar genomes, but controlled crosses between aspen and these neighboring species consistently fail. The reason has been unknown for decades. Working from the haplotype-resolved aspen genome her lab produced (Zhou et al., 2023, _Plant Journal_, cover story) — one of the first genomic assemblies to accurately map highly repetitive sequences — Dr. Chung-Jui Tsai and her team identified a candidate explanation: roughly seven percent of the aspen genome consists of a single repeating DNA unit, called M147, organized in tandem arrays that stretch across millions of base pairs per chromosome. This sequence is absent in every other _Populus_ section, and the single exception — a tropical aspen that lacks M147 — is also the one aspen that can occasionally hybridize across sections. A second aspen-exclusive feature, a variant of CENH3 (a protein essential to chromosome function during cell division), co-occurs with M147 in a pattern too consistent to dismiss as coincidence. The proposed NSF research pursues the hypothesis that M147 arrays and CENH3.2 interact to enforce reproductive isolation — a finding that would reframe satellite DNA as a functional actor in species-level evolution and open new possibilities for _Populus_ breeding. The implications extend beyond trees: satellite DNA appears throughout eukaryotic genomes including human chromosomes, where its functions remain poorly understood.

Satellite DNA Art Studio My role as co-PI centers on Activity 2 of the proposal's Broader Impacts plan — a component I designed and titled the _Satellite DNA Art Studio_. The scientific premise motivating this work is not merely that genomic data is difficult for non-specialists to access, but that the dominant modes of scientific data presentation — statistical tables, sequence alignment diagrams, heat maps — are themselves an epistemological problem. They are legible only to those already trained in their conventions, which means the most consequential genomic findings of the past decade have been effectively invisible to the public whose taxes funded them.

My research question within this collaboration is design-methodological: how can the principles and tools of graphic design be applied to genomic datasets not to make them decorative, but to make them genuinely intelligible — to translate structural complexity into visual form that communicates accurately to audiences outside the specialist community? This question sits at the intersection of communication design, data visualization, and the emerging field of science-art integration, and it is not rhetorical. Different visual representations of the same dataset produce measurably different comprehension outcomes. The problem of making M147 visible — literally: helping a viewer understand what it means that seven percent of a tree's DNA is a single repeating sequence — is a design problem with real epistemic stakes.

The proposed pedagogy integrates this research question directly into the graphic design curriculum. Under my supervision, students in Graphic Systems (ARGD 3020) — an upper-level course focused on visual problem-solving for complex, abstract content — would work with actual genomic datasets from the Tsai lab, developing original visualization approaches ranging from graphic simplification and information design to mixed media, three-dimensional installation, and generative AI. The student work functions not as illustration of pre-existing conclusions, but as a research catalog: multiple, competing visual translations of the same data, each representing a different design strategy, each legible (or not) to different audiences. That catalog is itself the scholarly output — a documented investigation of which visual approaches produce comprehension, and for whom.

A dedicated exhibition pipeline was designed to move the work into public space: campus galleries at the Lamar Dodd School of Art, the Georgia Museum of Art, public libraries, parks, and outdoor venues in Athens, with national exhibition as a longer-term trajectory. A public-facing website would serve as a permanent online documentation of both the science and the design process.

To develop the pilot version of this curriculum in advance of full funding, I was awarded a $1,000 mini-grant from the UGA Arts Collaborative. This grant is supporting the design of an intensive 2–3 day cross-disciplinary workshop, bringing together students from graphic design and the biological sciences to work collaboratively with satellite DNA data. The workshop is in active planning and represents the first operational phase of a curriculum that, at full scale, would run across a semester and produce an exhibitable body of work.

Peer Review Record The NSF proposal was reviewed by five external specialists and received a panel rating of Very Good across four of five reviews. All five reviewers addressed the Broader Impacts component explicitly; none raised criticisms. Two reviewers quoted the science-art integration as a specific strength for the Broader Impacts criteria. One wrote that "the team assembled to provide the teaching and training associated with this project in both biological sciences and the arts is compelling and likely to generate interest to students and the public." Another called working with a graphic design specialist "a very nice touch" and noted it as a genuine opportunity for graphic design students. A third described the activities as "well-considered" and "creative," specifically naming the inclusion of art students as a novel approach to public engagement.

The only substantive request across all reviews was the addition of audience metrics for future submissions — number of non-science majors reached, exhibition visitors, online gallery engagement. This is an addressable note that I am building into the documentation framework for the workshop and any subsequent curriculum implementation.

The scientific component was not funded due to insufficient preliminary data on the CENH3.2 hypothesis — a genetics question entirely outside my scope. My component received no criticism and the peer review record is on file.

---

Status and Trajectory

The grant is currently under revision. Dr. Tsai and Senior Research Associate Ran Zhou are leading the scientific revision; resubmission timeline is pending. My Broader Impacts design — the curriculum framework, exhibition pipeline, and pedagogy — is fully developed and independent of the grant timeline. The pilot workshop supported by the Arts Collaborative mini-grant proceeds in parallel, generating preliminary documentation and pedagogical data that will strengthen the next submission. When the grant is resubmitted, my component will include the audience metrics requested by reviewers and the documented outcomes of the pilot workshop as evidence of feasibility.

This project represents a methodological strand that runs across my research program: the conviction that design is not downstream of knowledge production, but constitutive of it. How data is represented determines what can be seen and understood. The _Satellite DNA Art Studio_ applies that conviction to one of the most technically inaccessible datasets in contemporary biology — and asks graphic design students to treat scientific incomprehensibility not as a limitation of the audience, but as a design problem waiting to be solved.

Data Visualization Pilot Workshop

The pilot workshop is the first operational phase of the _Satellite DNA Art Studio_, the Broader Impacts component of the NSF _Function and Dynamics of Aspen-Specific M147 Tandem Repeat Arrays_ proposal. Its purpose is to test, document, and refine a condensed version of the cross-disciplinary curriculum that Kappenstein is developing as co-PI, generating both pedagogical data and preliminary evidence of feasibility for the NSF revision.

The workshop brings together a small cohort of 6–10 students from graphic design and the biological sciences for three days of intensive, collaborative work with actual genomic datasets provided by the Tsai lab. It is structured as a research instrument as much as a pedagogical event: the animating question is not "can students make science visually appealing?" but "which design strategies produce genuine comprehension of complex genomic data in non-specialist audiences?" Each day advances that question in a different register.

Day 1 establishes shared ground. Biology and graphic design students encounter the same dataset — chromosome maps, M147 distribution diagrams, sequence abundance visualizations — from their respective disciplinary positions. A structured looking exercise makes the gap between scientific literacy and visual legibility explicit and productive: biology students articulate what the data shows; design students articulate what it communicates, and what it fails to communicate. That gap is the brief. Each participant or pair selects one aspect of the M147 dataset and a defined non-specialist audience, and produces first-concept sketches by end of day.

Day 2 is design development and structured critique. Participants work in their chosen medium — print, mixed media, digital, three-dimensional — without format constraint, translating their concept into visual form. A mid-day critique crosses the disciplinary divide deliberately: biology students evaluate scientific accuracy; graphic design students evaluate communicability. Points of disagreement between the two are treated as data, not error. Participants iterate based on the critique, and end the day with near-final work documented and brief written reflections collected. Those reflections constitute the first layer of pedagogical documentation for the NSF revision.

Day 3 moves to installation and public presentation. Work is installed as prototypes in the gallery space at the Lamar Dodd School of Art — resolved enough to exhibit, explicitly framed as first-iteration research rather than finished product. A small invited audience attends a brief public presentation in which participants explain their work and their design decisions. Professional documentation of the installed work is produced. A structured debrief closes the workshop, generating the facilitation notes, audience metrics, and participant reflection data that will form the evidence base for the full-scale curriculum proposal.

The $1,000 Arts Collaborative mini-grant supports materials, installation hardware, catering across the three working days, and documentation. All facilitation is provided by Kappenstein as in-kind contribution; data, datasets, and scientific expertise are provided by Dr. Tsai and Senior Research Associate Ran Zhou as in-kind from the Warnell School. Both contributions are being documented explicitly as evidence of institutional commitment for the NSF resubmission.

The workshop is currently in active planning. Its outputs — exhibited prototypes, participant reflections, audience metrics, and a documented exhibition record — will address the one substantive request made by NSF reviewers across all five reviews: the addition of measurable audience impact data to the Broader Impacts component. No reviewer criticized the _Satellite DNA Art Studio_ concept; two cited it as a specific strength. The pilot workshop transforms that concept into evidence.

• neuroline_01
• neuroline_02a
• neuroline_02b

Video installation work investigating pattern recognition, neural feedback loops, and the threshold between signal and noise. Two works exhibited; third in development.

The neuroline series investigates a specific perceptual phenomenon: the involuntary extraction of grids, rhythms, and patterns from surfaces, textures, and light. This is not a stylistic interest or an aesthetic preference — it is a documented feature of Prof. Kappenstein's neurotype, formally identified following her 2022 neurodivergence diagnosis. The research question the series poses is whether an experience that originates in a particular kind of mind can be translated into a spatial form legible to a different kind of mind.

The series works with two interlocking materials: the archives — a collection of analog and digital video, photography, screenshots, and screen recordings spanning more than four decades, with the earliest precisely dated footage from 1984 — and salvaged electronics sourced from UGA's surplus warehouse and elsewhere. The footage is not illustrative documentation of a concept. It is primary research data: recordings made across more than four decades by a mind that was already doing this, before there was language for what it was doing.

neuroline_01

neuroline_01 (C-U-B-E Gallery, Lamar Dodd School of Art, November 14 – December 15, 2023) was Prof. Kappenstein's first solo exhibition at UGA. The installation drew from the archives, assembled into a 30-minute loop projected in the gallery space. A second component introduced a live analog feedback mechanism: a camera filmed the projection screen and the people in front of it, with that live feed mixed directly into the projected image. Audience members could interact with the loop in real time by blocking part of the projection path — their presence altered the signal, making the viewer part of the pattern rather than a passive observer of it. The installation enacted, rather than illustrated, the research question: a perceptual loop that cannot be fully controlled, only inhabited.

neuroline_02a

neuroline_02a (Lupin Foundation Gallery, Lamar Dodd School of Art, February 8 – March 20, 2024) was presented as part of the “Trust Fall” Faculty Exhibition. The installation comprised three 6-minute video loops playing concurrently at slightly variable speeds, generating a combinatorial visual field that resists repetition. All hardware was constructed from electronics sourced from UGA's surplus warehouse and other salvaged components — a practice Prof. Kappenstein terms "Yestertech": a working studio philosophy in which decommissioned and obsolescent technology is treated not as constraint but as conceptual position. Using hardware that institutions have discarded enacts an argument about what remains useful, what remains valid, and what accumulates meaning precisely by persisting past its intended obsolescence.

neuroline_02b

neuroline_02b is the third installation in the series, currently in planning. The concept extends the Yestertech methodology to a new hardware form: an array of salvaged miniature screens — sourced from decommissioned digital cameras and similar devices — receiving simultaneous video feed. The technical feasibility of routing simultaneous feeds to multiple low-resolution, architecturally varied screens is under active investigation. The installation, if realized, would produce a fragmented, distributed image field: the same perceptual phenomenon rendered across surfaces that were never designed to work together.

The neuroline series is active. Two installations have been exhibited at the Lamar Dodd School of Art; a third is in planning. The research program it represents — translating neurodivergent perceptual experience into spatial form — is ongoing, and the archives from which it draws continue to grow.

neuroline emerges from a broader retrospective inquiry: a re-examination of 45 years of creative work made before Prof. Kappenstein had language for what she was investigating. A parallel inquiry is currently in development in the pareidolia photography series, which applies a similar retrospective approach to a distinct perceptual phenomenon.

Multi-phase research project: (1) viability/usability study for a unified UGA+ACC transit map; (2) AI-assisted readability optimization; (3) real-time interactive map; (4) AI route optimization. Phase 1 (UGA map draft) complete; Phase 2 (combined UGA+ACC) in progress.

Athens Transit Unified Map T2 Language & System · T3 Access & Justice COLLABORATORS** Interdisciplinary collaboration with mathematics and computer science in formation

Context

No unified map of the Athens public transit system currently exists. Riders navigate a fragmented landscape: separate maps for UGA bus routes, separate maps for Athens-Clarke County routes, no single view showing how the two systems connect. The most recent attempt at a comprehensive UGA bus map dates to 2015. This fragmentation is artificial. Both systems are fare-free and available to all riders; from a user perspective, they function as a single network. The UGA transit app includes ACC routes, but segregates them in a separate area of the interface — a design decision that reinforces a bureaucratic boundary invisible to the person standing at a bus stop.

The fragmentation is not only organizational. Both systems share a vendor platform, Passio Go, which renders routes as overlaid lines on a geographically accurate base map. Because the underlying map is geographically correct, routes that share corridors are drawn directly on top of one another, becoming illegible. A rider trying to understand the full network cannot see it — not because the information doesn't exist, but because the display system makes it unreadable. This is an accessibility problem at multiple levels: it fails riders who are unfamiliar with the city, riders with cognitive or visual processing differences, and any rider who needs to plan a multi-leg journey across both systems.

My research asks two related questions: is a unified, readable map of the full Athens transit system achievable, and can artificial intelligence be used to optimize both the map's design and the underlying route structure it depicts? The first question is a design research problem with direct community application. The second points toward a larger ambition: an agentic design system capable of maintaining, optimizing, and updating a transit map autonomously in response to real-world route changes.

My Contribution

I bring to this project a specific prior expertise in complex transit map design — the kind of systems-level cartographic problem that requires both data literacy and typographic precision. That expertise is the methodological foundation of the research, not its claim.

The central design problem is geometric: geographically accurate maps cannot display overlapping routes legibly. The solution, established in the tradition of transit cartography from Beck's 1933 London Underground map forward, is schematization — a designed abstraction that sacrifices geographic literalism in favor of relational clarity. Applying schematization to the combined Athens network, with its irregular street grid, shared corridors, and divergent route structures, required sustained iterative problem-solving across variables that include line weight, angle constraints, corridor sequencing, braiding order, node spacing, and typographic hierarchy.

My process moved through four documented stages. Initial route structure and spatial logic were worked out through large-format hand drawings, allowing rapid testing of layout alternatives before committing to software. Structural variables were then developed and refined across multiple iterations in Figma. Once the visual system was stable, production shifted to Affinity Designer, selected for its superior vector precision for this class of problem. The near-complete artifact, v6.02, is a unified map of the full Athens transit network — both UGA and Athens-Clarke County routes — at approximately 30 × 40 inches, representing the first complete map of the combined system. At this scale the map is legible as a large-format print; a paper map is a useful byproduct, not the end goal.

The current phase of the research addresses a problem this project shares with transit map design globally: maintaining map currency as routes change. Larger rail systems have addressed live vehicle tracking through electronic displays and GPS-based mobile apps, but no established solution exists for route changes in bus-based systems. Two distinct technical problems must be solved. First, live GPS tracking data must be translated onto a map that is deliberately non-geographic — the schematic map must be "mapped" against the real network it abstracts. Second, and more complex, an AI system must be able to reference not only route and stop data as GPS coordinates, but the full bus-navigable street network, and make layout decisions — sequencing, braiding, stop placement, line ordering — on demand, for any route change. I am currently conducting early experimental work on this problem using generative AI tools in Figma, working with live data from both transit systems. A collaboration with researchers in mathematics and computer science is in formation.

Status and Trajectory

Version 6.02 of the unified Athens transit map is near-complete — the first complete map of the combined UGA and Athens-Clarke County bus network. The paper map is a documented research artifact and a proof of concept; the research goal is an agentic design system that can maintain and update a schematic transit map autonomously, reading live route data and making layout decisions — sequencing, braiding, stop placement — on demand. Active experimental work toward this goal is underway now, using generative AI tools in Figma with live data from both transit systems. The formation of an interdisciplinary collaboration with mathematics and computer science is in progress, which will be necessary to advance the AI optimization and route redesign phases of the research program. The questions at the center of this project — whether design-based methods can optimize the infrastructure they visualize, and whether an agentic system can sustain a living map — are positioned for external funding in both design and computer science.

• Hewitt Multidimensional Perfectionism Scale Study
• "Mindfulness and the Creative Process" — AIGA Design Conference 2023

A multi-year study correlating visual creative outputs with perfectionism trait profiles, developed from ARGD 2010 studio data. Includes a dataset from Spring 2023 and a national conference presentation.

Hewitt Multidimensional Perfectionism Scale Study

Drawing on two decades of design teaching and my yoga instructor training, I developed a repeatable research methodology pairing a validated perfectionism instrument with a standardized meditative drawing task to test whether perfectionism and creativity operate in measurable opposition.

Context

Across two decades of teaching graphic design and a 500-hour yoga instructor training (RTY-500, E-RYT200), I have observed a consistent inverse relationship between perfectionism and creative output — that the closer a designer moves toward an ideal of perfection, the more constrained creative space becomes. Perfectionism and creativity, I argue, operate in direct opposition: perfectionism narrows; creativity demands openness.

This research asks whether that relationship is measurable. The methodology pairs two instruments: the Hewitt Multidimensional Perfectionism Scale (MPS), a validated psychological questionnaire measuring self-oriented, other-oriented, and socially prescribed perfectionism; and a standardized drawing task administered immediately following a guided gazing meditation. The gazing instrument is a sheet of paper with a central dot and an approximately seven-inch circle — a form with deep roots in Vedic contemplative practice as an object of concentrated attention. Participants gaze at the dot, pen in hand, and on command draw freely within the circle. A small, unobtrusive identification number in the corner of each sheet links the resulting drawing to its corresponding MPS response, creating a paired dataset of perfectionism profiles and visual output produced under conditions of deliberate openness.

Contribution

I designed the methodology pairing a validated perfectionism instrument with a standardized drawing task administered in a meditative state. The research module distills the pedagogical framework developed in the Summer 2021 course into a repeatable, controlled procedure: participants complete the Hewitt MPS, undergo a guided gazing meditation on a standardized visual instrument — a sheet bearing a central dot within an approximately seven-inch circle — and draw freely within the circle. The same instrument is used in every administration; what varies is the participant, not the task. A small identification number printed in the corner of each sheet links each drawing to its corresponding MPS response.

I administered the module in two contexts. In Spring 2023, I integrated it into ARGD 2010 (Graphic Design Survey) at UGA, producing 69 MPS responses and 65 paired drawings. (Four participants completed the MPS but did not produce a drawing, having arrived late or not followed instructions.) In October 2023, I presented the research at the AIGA Design Conference in New York City — the primary annual gathering of the professional design community in the United States — and conducted a workshop using the same methodology. The AIGA session was selected through competitive proposal review, constituting external validation of the research. The workshop produced 43 MPS responses and 43 paired drawings. The combined dataset comprises 112 MPS responses and 108 drawings, each anonymized and linked by individual identification number.

Research-as-Teaching

"Mindfulness and the Creative Process" originated in the classroom. The Summer 2021 course — ARGD 4080 / ARST 4915 / ARST 6915, cross-listed for undergraduate and graduate students — was the generative environment in which the methodology was first tested. Over four weeks, seven students worked at the intersection of design practice and contemplative tradition: sitting and walking meditations, intuitive drawing, translation of sound into visual form, exploration of sacred geometry as both design object and meditative instrument. The course did not measure anything. It asked students to become aware of their own creative process — its conditions, its resistances, its requirements.

The stand-alone research module emerged from that course as a distillation: what could be isolated, controlled, and administered repeatably? The gazing meditation and drawing task retained the pedagogical core — deliberate suspension of evaluative pressure — while adding the Hewitt MPS to make the question empirical. Teaching generated the research question; the research module is now redeployed in the teaching context (ARGD 2010) and tested beyond it (AIGA Design Conference).

The student population in ARGD 2010 is not a neutral sample. DRC notifications and student self-disclosure indicate a notable proportion of neurodivergent students in the courses where the module has been administered. This is not incidental. Perfectionism is a documented feature of AuDHD presentation, and the course population that generated the dataset is the same population for whom the research question carries the most direct stakes.

Status and Trajectory

The dataset is complete and documented: 112 MPS responses paired with 108 drawings, collected across two administrations (ARGD 2010, Spring 2023; AIGA Design Conference, October 2023). Analysis is pending. The open research question — whether perfectionism profile predicts measurable characteristics in the resulting drawings — remains unanswered.

The project sits at the intersection of two research threads. As a T1 inquiry, it asks how psychological orientation shapes the visual forms a designer produces under conditions of deliberate openness. As an emergent T3 concern, it asks what that answer would mean for neurodivergent designers — a population in which perfectionism is a known barrier to output and for whom the stakes of that question are not abstract. The 2022 AuDHD diagnosis sharpened this trajectory: what began as a universal claim about creativity and perfectionism is now legible as a question with particular relevance to neurodivergent creative practice.

The methodology's next phase depends on analysis of the existing dataset. The advent of AI-assisted pattern recognition across visual datasets opens a possible path: if perfectionism profile correlates with identifiable visual characteristics in the drawings, AI tools may make that correlation detectable at scale. This is a trajectory, not a claim. The dataset exists. The analysis has not yet been conducted.

"Mindfulness and the Creative Process" — AIGA Design Conference 2023

The October 2023 workshop at the AIGA Design Conference in New York City brought the methodology to a professional audience of designers, educators, and researchers — a population distinct from the undergraduate students who generated the first dataset. Forty-three participants completed the Hewitt MPS, underwent the guided gazing meditation, and produced drawings within the standardized instrument. The workshop format tested portability: could the methodology hold outside the institutional structure of a course, with self-selecting professionals rather than enrolled students? The 43 paired responses suggest it can. [insert participant feedback quotes here] The session was selected through AIGA's competitive proposal review process, and the dataset it generated constitutes the second of two research administrations, complementing the Spring 2023 ARGD 2010 collection.

• Exhibition Concept

A typeface as political instrument: Verboten examines what happens when a font does the compliance for you.

Exhibition Concept

Typeface design and installation concept · 2025–present · Sole creator · verbotenfont.org

Context

In 2025, the federal government prohibited the use of specific research terms across agencies including the CDC and NIH — words and phrases associated with diversity, equity, inclusion, and related areas of scientific inquiry. The prohibition operates at the level of language itself: researchers and institutions are expected to remove these terms from documents, grant applications, and public-facing materials, often without explicit mandate. The question _Verboten_ poses is typographic: what happens when a font performs that censorship on your behalf?

Kappenstein's Contribution

_Verboten_ is a typeface built on a modified version of Arial — one of the most ubiquitous, institutionally neutral typefaces in existence. Using OpenType ligature substitution, Prof. Kappenstein has encoded each term from the 2025 federal prohibited list as a ligature: when any of these words or phrases is typed in _Verboten_, the font automatically replaces it with a crossout glyph that renders the word as if struck through with a black marker. The redaction is not applied by a human hand or an external tool. It is built into the letterforms. The font does the compliance for you.

The technical method is precise and conceptually loaded. OpenType ligature substitution is a standard mechanism in professional type design, used to handle character combinations that require special treatment — ligatures like _fi_ or _fl_, or contextual alternates in Arabic script. Here, the same mechanism is repurposed: instead of resolving a typographic problem, it performs a political act. The typeface enacts what theorist Timothy Snyder calls anticipatory obedience — the preemptive compliance with authority that makes suppression self-sustaining. _Verboten_ makes that compliance visible, material, and legible as a designed object.

The project extends into installation. The proposed public form places a networked printer on a pedestal. Visitors upload a document to verbotenfont.org; an automated system inspects the file, resets the text in _Verboten_, generates a PDF, returns it to the sender by email, and simultaneously prints it on the gallery printer. The output accumulates on the floor. Pages may be pinned to the wall. The printer is the performer; the falling paper is the evidence. No human intervention is required. The system runs.

_Verboten_ is developed in FontLab. The Regular weight is near-complete. Italic, Bold, and Bold Italic weights are in development. The banned terms list, which is itself subject to ongoing federal revision, functions as a living dataset — its maintenance is part of the research.

Peer Review / Reception

_Verboten_ was developed for inclusion in a Lamar Dodd School of Art faculty exhibition; the show was canceled before the work could be realized. A proof-of-concept installation is scheduled at the C-U-B-E Gallery, Lamar Dodd School of Art, University of Georgia, summer 2026. The project's public record is documented at verbotenfont.org.

Status and Trajectory

_Verboten_ is an active project in development. The Regular weight typeface is near-complete; Italic, Bold, and Bold Italic weights are in progress. A proof-of-concept installation is scheduled at the C-U-B-E Gallery for summer 2026, following the completion of the academic year. The web interface and networked printer output will be realized in that context. As the federal list of prohibited terms continues to expand, the typeface requires ongoing curatorial maintenance — a condition that is, itself, part of the research record.

• Faces

A photography project investigating pareidolia — the perceptual tendency to find meaningful patterns, faces, and forms in random or ambiguous visual data. Strong visual documentation confirmed.

Faces

More to come.

• The FruitFont Project
• Typeface Design — AI-Assisted Variation
• Environmental Branding — AI-Assisted Ideation
• Reverse Engineering a Visual System — ARGD 3020
• Creating with AI — ARGD 4080 / FYOS Co-Pilot

Systematic investigation of generative AI as design tool, pedagogical challenge, and object of critical inquiry — developed across five courses at UGA, deployed in live studio environments, and generating a multi-year comparative dataset.

The FruitFont Project

ARGD 4080 / ADPR 5990, Summer 2023 and ongoing. Students design complete alphabets whose formal characteristics derive from an assigned fruit: the matte texture and taper of a banana peel, the saturated gradients and seed geometry of a raspberry, the sectional structure of a pomegranate. Repeated across successive course offerings, the project has accumulated a multi-year dataset comparing outputs across cohorts and tracking how quality shifts — or doesn't — as AI tools become embedded in student practice.

Typeface Design — AI-Assisted Variation

ARGD [course number]. Students design a typeface, then develop AI-assisted variants using the original as a style reference. More to come.

Environmental Branding — AI-Assisted Ideation

ARGD [course number]. AI integrated into the conceptual development phase of an environmental branding project. More to come.

Reverse Engineering a Visual System — ARGD 3020

Active, Spring 2026. Students select an architectural style, design movement, or philosophical framework and develop a singular master composition from primary library research. They extract five formal rules from that composition, then generate derivative panels both manually (Step 6) and using generative AI — Adobe Firefly and one additional model of their choice (Step 7). All prompts, iterations, and process documentation are collected alongside the final outputs. The parallel dataset — human-derived and AI-generated, across 40+ students, following identical rule sets — is the research object. This is the first systematic, course-scale study of AI versus human rule-following in visual system generation conducted at Lamar Dodd.

Creating with AI — ARGD 4080 / FYOS Co-Pilot

Upper-division studio (ARGD 4080) and first-year seminar (FYOS Co-Pilot). Paired offering examining AI as creative collaborator at two points in the design curriculum simultaneously — early formation and advanced practice. More to come.

• Banned Books Protest Posters
• Swami Jaya Devi — Book Cover
• Dodd Wayfinding System
• Queer Bibliography 2026

Design practice engaged with social and institutional contexts — protest graphics, book design, and wayfinding at civic scale.

Banned Books Protest Posters

Indivisible Georgia, January 2026. Twelve posters designed for front-and-back wear as sandwich boards at a Banned Books protest in Athens, Georgia. Documentation exists.

Swami Jaya Devi — Book Cover

Book cover design, completed 2022. More to come.

Dodd Wayfinding System

Comprehensive wayfinding and environmental signage system for the Lamar Dodd School of Art building. See also: Service section for full institutional context and scope.

Queer Bibliography 2026

Co-designed poster and live letterpress printing workshop at the Queer Bibliography 2026 conference, March 12–14, 2026. The workshop was hosted by Lamar Dodd School of Art and co-sponsored by the Mellon Society of Fellows and the Bibliographical Society of America. Co-organized with Eileen Wallace (Lamar Dodd School of Art).

The research arc: generative AI output set by hand in letterpress — at a queer bibliography conference, in the current political climate. The AI-to-letterpress pipeline inverts the usual direction of design technology, using a high-fidelity manual process to interrogate the outputs of a generative one.

The teaching & service dossier is forthcoming.

The teaching & service dossier is forthcoming.