EMDR Bilateral Stimulation Research Platform Features¶

Last Updated: November 11, 2025 Project Version: Research Platform v1.0 Hardware: Dual Seeed XIAO ESP32-C6 Framework: ESP-IDF v5.5.0 via PlatformIO

Overview¶

This document describes the research platform capabilities of the EMDR bilateral stimulation device, which extends beyond standard therapeutic parameters to enable scientific investigation of bilateral stimulation mechanisms.

Dual Purpose: 1. Clinical Tool: Standard EMDR therapy (0.5-2 Hz, fixed patterns) 2. Research Platform: Extended parameters (0.25-2 Hz, multiple patterns)

Research Platform Architecture¶

BLE Service Structure¶

Configuration Service (Mobile App Control) - UUID: 6E400002-B5A3-F393-E0A9-E50E24DCCA9E - Purpose: User interface for parameter configuration

Bilateral Control Service (Device-to-Device) - UUID: 6E400001-B5A3-F393-E0A9-E50E24DCCA9E - Purpose: Real-time bilateral coordination

Stimulation Patterns¶

1. BILATERAL_FIXED (Standard EMDR)¶

Traditional bilateral stimulation with fixed motor directions:

Time (ms)    Server Device    Client Device
0-250        FORWARD ON       OFF
250-500      OFF             REVERSE ON
500-750      FORWARD ON       OFF
750-1000     OFF             REVERSE ON

Applications: - Standard EMDR therapy - Baseline measurements - Control condition

Research Questions: - Optimal frequency for different conditions - Relationship between frequency and processing speed

2. BILATERAL_ALTERNATING (Research Mode)¶

Both devices alternate motor direction each cycle:

Time (ms)    Server Device        Client Device
0-250        FORWARD ON          OFF
250-500      OFF                REVERSE ON
500-750      REVERSE ON         OFF
750-1000     OFF               FORWARD ON
1000-1250    FORWARD ON         OFF
1250-1500    OFF               REVERSE ON

Applications: - Direction preference studies - Habituation reduction research - Motor adaptation studies

Research Questions: - Does alternating direction improve outcomes? - Impact on sensory habituation - User preference for direction patterns

3. UNILATERAL (Control Studies)¶

Single device operation for research controls:

Time (ms)    Server Device    Client Device
0-250        ON              OFF (always)
250-500      OFF             OFF (always)
500-750      ON              OFF (always)
750-1000     OFF             OFF (always)

Applications: - Control conditions - Baseline comparisons - Single-sided stimulation studies

Research Questions: - Bilateral vs unilateral effectiveness - Necessity of bilateral alternation - Hemispheric processing differences

Extended Parameter Ranges¶

Frequency Range: 0.25-2 Hz¶

Frequency	Cycle Time	Classification	Research Application
0.25 Hz	4000ms	Ultra-slow	Dissociative disorders, slow processing
0.5 Hz	2000ms	Slow	Standard EMDR minimum, elderly populations
0.75 Hz	1333ms	Moderate-slow	Attention disorders
1.0 Hz	1000ms	Standard	Typical therapeutic rate
1.25 Hz	800ms	Moderate-fast	Enhanced processing
1.5 Hz	667ms	Fast	Anxiety disorders
1.75 Hz	571ms	Very fast	ADHD applications
2.0 Hz	500ms	Ultra-fast	Standard EMDR maximum

Motor Intensity: 30-80% PWM¶

Safety Constraints: - Minimum 30%: Ensures perceptible stimulation - Maximum 80%: Prevents motor damage and tissue irritation

Research Applications: - Sensory threshold studies (30-40%) - Standard therapy (40-60%) - Enhanced stimulation (60-80%)

Duty Cycle: 10-50%¶

Duty Cycle	On Time (1Hz)	Application
10%	50ms	Micro-stimulation research
15%	75ms	Minimal effective dose studies
25%	125ms	Standard therapy (battery efficient)
35%	175ms	Enhanced stimulation
50%	250ms	Moderate duty (battery efficient)
75%	375ms	High intensity stimulation
100%	500ms	Maximum duty (uses entire ACTIVE period, respects INACTIVE period)
nNote: All duty levels guarantee motor is OFF for at least 50% of total cycle time. Duty cycle is percentage of ACTIVE period only.

Session Duration: 20-90 minutes¶

20 minutes: Minimum research protocol
30 minutes: Short therapy session
45 minutes: Standard therapy session
60 minutes: Extended therapy session
90 minutes: Maximum research session

Research Protocol Examples¶

Protocol 1: Direction Preference Study¶

Objective: Determine if motor direction affects therapeutic outcome

Design: Within-subjects, counterbalanced

Procedure:

1. Baseline Assessment (5 min)
   - BILATERAL_FIXED at 1 Hz, 40% PWM

2. Condition A (10 min)
   - BILATERAL_FIXED pattern
   - Record subjective ratings

3. Rest Period (5 min)
   - No stimulation

4. Condition B (10 min)
   - BILATERAL_ALTERNATING pattern
   - Record subjective ratings

5. Post Assessment (5 min)
   - Compare conditions

Protocol 2: Minimal Stimulation Threshold¶

Objective: Determine minimum effective stimulation parameters

Design: Ascending method of limits

Procedure:

1. Start: 10% duty cycle, 30% PWM, 1 Hz
2. Increment: +5% duty cycle every 5 minutes
3. Measure: Subjective perception threshold
4. Continue: Until therapeutic effect reported
5. Verify: Repeat at identified threshold

Protocol 3: Frequency Optimization¶

Objective: Find optimal frequency for specific populations

Design: Randomized blocks

Procedure:

Day 1: 0.5 Hz (20 min)
Day 2: 1.0 Hz (20 min)
Day 3: 1.5 Hz (20 min)
Day 4: 2.0 Hz (20 min)
Day 5: 0.25 Hz (20 min) - research extension

Measure: Processing speed, subjective comfort, therapeutic outcome

Data Collection¶

Real-Time Parameters¶

Logged every stimulation cycle:

typedef struct {
    uint32_t timestamp_ms;        // Millisecond timestamp
    uint16_t commanded_cycle_ms;  // Target cycle time
    uint16_t actual_cycle_ms;     // Measured cycle time
    uint8_t motor_intensity;      // PWM percentage (30-80)
    uint8_t duty_cycle;           // On-time percentage (10-50)
    uint8_t pattern_type;         // FIXED/ALTERNATING/UNILATERAL
    uint8_t device_role;          // SERVER/CLIENT
    uint16_t sequence_number;     // For packet loss detection
    uint8_t battery_percent;      // Battery level
    int16_t back_emf_mv;         // Motor back-EMF (if enabled)
} research_data_t;

Session Metadata¶

Stored at session start/end:

typedef struct {
    uint32_t session_id;          // Unique session identifier
    uint32_t start_time;          // Unix timestamp
    uint32_t duration_ms;         // Total session length
    uint8_t protocol_version;     // Research protocol ID
    uint16_t total_cycles;        // Number of bilateral cycles
    uint16_t packet_loss_count;   // BLE reliability metric
    uint8_t pattern_changes;      // Number of pattern switches
} session_metadata_t;

Safety Features¶

Hard Limits (Cannot Override)¶

PWM Range: 30-80% (enforced in firmware)
Non-overlapping: Devices never stimulate simultaneously
Emergency Stop: 5-second button hold always active
Watchdog Timer: 2-second timeout prevents lockups

Soft Limits (User Warnings)¶

Session > 60 min: "Extended session - confirm to continue"
Duty > 40%: "High duty cycle may cause motor heating"
Frequency < 0.5 Hz: "Below standard EMDR range"
PWM > 70%: "High intensity - monitor for comfort"

Research Mode Indicators¶

LED patterns indicate research vs clinical mode:

Clinical Mode: Solid green during operation
Research Mode: Pulsing blue during operation
Parameter Change: Brief yellow flash
Safety Limit: Red flash with warning

BLE Characteristic Reference¶

Bilateral Control Service Characteristics¶

Characteristic	UUID Suffix	Type	Range	Access
Bilateral Command	0101	uint8	0-6	Write
Total Cycle Time	0201	uint16	500-4000ms	R/W
Motor Intensity	0301	uint8	30-80%	R/W
Stimulation Pattern	0401	uint8	0-2	R/W
Device Role	0501	uint8	0-2	Read
Session Duration	0601	uint32	20-90 min	R/W
Sequence Number	0701	uint16	0-65535	Read
Emergency Shutdown	0801	uint8	1	Write
Duty Cycle	0901	uint8	10-50%	R/W

Implementation Status¶

Completed Features¶

✅ Basic bilateral coordination (BILATERAL_FIXED)
✅ BLE Configuration Service
✅ Standard EMDR frequencies (0.5-2 Hz)
✅ Motor PWM control (30-90%)
✅ Battery monitoring
✅ Session timing

In Development¶

🚧 BILATERAL_ALTERNATING pattern
🚧 Extended frequency range (0.25 Hz)
🚧 Research data logging
🚧 Bilateral Control Service

Future Features¶

⏳ Real-time data streaming
⏳ Physiological sensor integration
⏳ Machine learning optimization
⏳ Cloud data synchronization

Comparison with Clinical Standards¶

Parameter	EMDR Standard	Research Platform	Rationale
Frequency	0.5-2 Hz	0.25-2 Hz	Explore slow processing
Pattern	Fixed bilateral	Fixed/Alternating/Uni	Study mechanisms
Intensity	Not specified	30-80% PWM	Safety + variability
Duty Cycle	Not specified	10-50%	Energy + comfort
Session	60-90 min typical	20-90 min	Research flexibility

Research Ethics Considerations¶

Required for Human Studies¶

IRB approval for research protocols
Informed consent with research disclosure
Data anonymization (no PII stored)
Right to withdraw from research
Clear clinical vs research mode indication

Device Labeling¶

"Research features are investigational"
"Standard EMDR mode for clinical use"
"Extended parameters for research only"

Future Research Directions¶

Phase 2: Sensor Integration¶

Heart rate variability (HRV)
Galvanic skin response (GSR)
EEG correlates
Eye tracking

Phase 3: Closed-Loop Control¶

Adaptive frequency based on HRV
Intensity modulation via GSR
Pattern switching on habituation
Personalized protocols

Phase 4: Multi-Site Studies¶

Standardized research protocols
Cloud data aggregation
Cross-population comparisons
Efficacy meta-analysis

Phase 5: AI Optimization¶

Machine learning parameter selection
Predictive therapeutic outcomes
Personalized treatment protocols
Real-time adaptation algorithms

References¶

Shapiro, F. (2001). Eye Movement Desensitization and Reprocessing (EMDR): Basic Principles, Protocols, and Procedures.
EMDRIA Standards: https://www.emdria.org
ESP-IDF Documentation: https://docs.espressif.com/projects/esp-idf/
NimBLE Documentation: https://mynewt.apache.org/latest/network/

Contact & Contributing¶

Repository: https://github.com/lemonforest/mlehaptics License: GPL v3 (Software), CERN-OHL-S v2 (Hardware) Research Collaboration: Open to academic partnerships

This document describes research capabilities that extend beyond standard EMDR therapy. Research features should only be used in appropriate research contexts with proper oversight and consent.