BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251016T135127Z
LOCATION:Grand Hall L
DTSTART;TZID=America/Chicago:20251014T165000
DTEND;TZID=America/Chicago:20251014T171000
UID:HFESAM_ASPIRE 2025_sess171_LECT595@linklings.com
SUMMARY:Adaptive De-escalation Trainer: Piloting a RAG-Enhanced, Emotional
 ly Modulated AI Simulator for Police Training
DESCRIPTION:Eshwara Prasad Sridhar, Jose Lopez, Mohammad Islam, and Shuchi
 snigdha Deb (University of Texas at Arlington)\n\nEffective police de-esca
 lation training requires realistic practice, often limited by traditional 
 method’s scalability and consistency. We developed an adaptive AI simulato
 r enabling officers to practice de-escalation techniques across varied sce
 narios representing common crisis encounters, such as mental health issues
  or suicidal ideation. Our system integrates a large language model (LLM) 
 with retrieval-augmented generation (RAG) for contextual scenario groundin
 g. It employs dynamic Text-to-Speech (TTS) synthesis to modulate the AI's 
 emotional prosody. This emotional expression adapts based on keyword analy
 sis ("success"/"failure" phrases) of officer speech and an internal state 
 model, creating a responsive conversational partner. Our ongoing study wit
 h active police officers (target N=5-10) evaluates the system's feasibilit
 y and effectiveness. Through participant surveys (assessing interaction qu
 ality, emotional realism, workload via NASA-TLX) and system performance lo
 gs (latency, state changes, keyword impact), we investigate perceived conv
 ersational coherence/responsiveness (H1), perceived emotional realism (H2)
 , feedback loop validation (H3), and technical viability (H4). This invest
 igation evaluates the potential of adaptive conversational AI agents as sc
 alable, on-demand supplementary resources for law enforcement skills train
 ing. These systems may enhance officer confidence and communication strate
 gies by offering a secure environment for repeated practice. Concurrently,
  this work aims to identify limitations and areas requiring further develo
 pment within this AI-based pedagogical model.\n\nTrack: Aerospace Systems,
  Cognitive Engineering & Decision Making, General Sessions, Human AI Robot
  Teaming (AI)\n\nSession Chair: Elizabeth Veinott (Psychology and Human Fa
 ctors, Michigan Technological University)\n\n
END:VEVENT
END:VCALENDAR