Founding Product Designer with a PhD in Cognitive Psychology and behavioral research since 2006. Built the design language system for an AI-powered SaaS platform. Published on judgment under uncertainty; dissertation in psycholinguistics on word recognition. Those are the two mechanisms underlying defaults, naming, information architecture, and onboarding: the parts of a complex product where comprehension is the hard problem.
EXPERIENCE
SideKix HQ, Inc.
Founding Product Designer
June 2026–present
Design lead for an AI-powered SaaS platform; represent the design pod (up to five junior designers) in executive planning and roadmap decisions
Oversee project workflows and run design critiques while maintaining a full-stack design workload
Partner with engineering and product to define interaction patterns and component specs for a platform in active development
Product Designer
April 2026–June 2026
Led the 0-to-1 product design for the platform
Built a scalable Design Language System spanning tokens, components, iconography, type scale, color palette, and motion
Audited the mobile app against WCAG 2.1 AA; refined component specs and remediation priorities
Product Design Intern
March 2026–April 2026
Presented research to executives; findings shaped the roadmap on membership pricing, learning modules, and onboarding
Contributed to product design work including wireframing, user flow mapping, & component development in Figma
The Pennsylvania State University
Assistant Teaching Professor & Laboratory Director
July 2017–December 2025
Principal investigator on multiple end-to-end research streams examining how people process, interpret, & respond to information; designed and ran controlled experiments with 3–5 research assistants and ~200 participants per semester
Published peer-reviewed research on cognitive bias and decision-making with direct implications for UX design patterns
Translated complex datasets into interaction design, information architecture, and usability decisions; presented and defended findings to high-level stakeholders and collaborators
Directed research operations for the department’s participant pool; recruitment & scheduling for 35+ assistants and ~500 participants per semester, with data governance and confidentiality protocols throughout
Taught, advised, and mentored ~2,500 undergraduate and graduate students
Helped develop an interdisciplinary Behavioral Finance Certificate with business faculty (2020–2022), applying research on cognitive bias to financial decision-making
Arkansas Tech University
Assistant Professor
July 2015–May 2017
Principal investigator on mixed-methods experimental studies; managed participant recruitment and scheduling for the departmental research pool
Taught, advised, and mentored ~1,000 undergraduate and graduate students
Additional teaching appointments
Instructor of Record
University at Buffalo (SUNY; August 2009–July 2015; taught ~1,000 students including a 450-seat introductory lecture), Niagara County Community College (May 2014–June 2015), Daemen University (August 2010–July 2013)
EDUCATION
Professional Certificate, UX Design, 2026
Google on Coursera (credential ID: MBN0NAW4TZC2)
PhD, Cognitive Psychology, 2015 · MA, Psychology, 2011 · BA, Psychology, 2006
University at Buffalo (SUNY) — graduate research in psycholinguistics, 2006–2015; laboratory manager, 2009–2015, training and supervising the lab's research assistants
Dissertation topic: On word recognition and lexicalization, or how people perceive, encode, and learn new words under lexical competition from existing vocabulary (mental models, vocabulary acquisition, learnability, & onboarding)
CORE COMPETENCIES
Research: Mixed-methods user research (qualitative & quantitative), user interviews, observational studies, usability testing, surveys & card sorting, A/B testing, behavioral telemetry, personas, competitive audits
Design: wireframing, prototyping, storyboarding, UI fundamentals, information architecture & navigation design, design systems (tokens, auto-layout), affinity & journey mapping
Accessibility: WCAG 2.1 AA audits & accessibility standards
TOOLS & TECHNOLOGIES
Design & Collaboration: Figma, Framer, Claude Design, Miro, Adobe Premiere, Adobe Photoshop
Research & Data: SPSS, Qualtrics, REDCap, R
AI & Creative: Claude, Figma Make, Google Gemini & Flow, Canva, Adobe Firefly
Certification: Coursera AI-Powered Design & Creative Tools (2026; credential ID: KCIOLCUMU8AX)
RESEARCH WITH UX IMPLICATIONS
Anchoring effects and the influence of source credibility on judgment under uncertainty
Under review at Psychological Reports (Berg, S. A., 2026)
Source credibility moderated the anchoring effect: estimates spread far more widely between low and high anchors when the source was credible than when it was not
Implicit cues about the source reduced the bias where an explicit instruction to disregard had only partially reduced it
Connects anchoring susceptibility to cognitive accessibility and choice architecture: interface quality as a credibility cue, pre-filled defaults as sludge, higher baseline cognitive load under assistive technology
Direct implications for trust signaling and AI-assisted decisions: expert endorsements, verified badges, and algorithmic recommendations amplify anchoring rather than merely inform it
Anchoring and judgment bias: Disregarding under uncertainty
Psychological Reports (Berg, S. A., & Moss, J. H., 2022) [doi.org/10.1177/00332941211016750]
Instruction to disregard an anchor narrowed the spread between anchors but did not remove the bias
Correction was asymmetric: subjects discounted the implausible anchor but not the plausible one
A caveat weakens a reference value without removing it; direct implications for price anchoring, pre-filled defaults, and form-field priming
Novel forms in the adult mental lexicon: Listening to new neighbors
ProQuest Dissertations & Theses Global (Berg, S. A., 2015) [Publication No. 3725897]
Newly learned word forms built to compete with familiar ones (“cathedruke” against “cathedral”) produced facilitation rather than the predicted interference; they primed shared sounds before they were learned well enough to compete lexically
Recognition was slower and less accurate for words with more similar-sounding neighbors; implications for menu labels, feature names, and onboarding terminology
