From Clinical Interview to Behavioural, Neurophysiological, and Sensor-Based Measures

Psychiatric diagnosis has always been probabilistic rather than absolute. Unlike conditions diagnosed by a single biomarker, disorders such as ADHD emerge from patterns of behaviour, cognition, and neurobiology, each influenced by context, development, and compensation.

The most reliable diagnostic strategy, therefore, is not reliance on a single tool—but serial, multimodal integration.

Modern ADHD assessment is increasingly moving toward a layered model that combines:

1. Clinical interview

2. Behavioural testing (CPT)

3. Neurophysiological assessment (QEEG)

4. Sensor-based measures such as eye tracking and motion tracking

Each layer adds signal while reducing noise.

The Baseline Reality: Limits of Clinical Interview Alone

The psychiatrist-led clinical interview remains the cornerstone of diagnosis. It establishes:

• developmental onset,

• cross-situational impairment,

• subjective distress,

• and differential diagnoses.

However, psychiatry has long acknowledged the limits of interview-only diagnosis.

Large DSM field trials and reliability studies show:

• Inter-rater reliability (kappa) for many psychiatric diagnoses ranges from 0.40–0.60 (moderate agreement).

• For adult ADHD, diagnostic agreement based primarily on interview and history typically falls around 60–70%.

These limitations arise from:

• recall bias (especially in adults),

• masking in high-functioning individuals,

• symptom overlap with anxiety, depression, sleep disorders,

• subjective reporting by parents, teachers, or patients.

The interview defines the clinical hypothesis. Objective tools are required to test it.

CPT: Quantifying Attention and Impulsivity

The Computerised Continuous Performance Test (CPT) adds objective behavioural measurement.

Across multiple studies and meta-analyses:

• CPT measures show moderate to large effect sizes (Cohen’s d ≈ 0.6–0.9) in distinguishing ADHD from controls.

• Sensitivity typically ranges from 65–85%, depending on task design and population.

• CPT is particularly sensitive to deficits in sustained attention and response variability—core ADHD features.

However, CPT performance can be influenced by:

• anxiety,

• fatigue,

• motivation,

• test familiarity.

Which is why CPT improves accuracy only when interpreted in clinical context.

QEEG: Adding Neurophysiological Evidence

Quantitative EEG (QEEG) introduces a neurobiological dimension.

Large normative-database studies report:

• ADHD-associated EEG patterns with sensitivities of approximately 70–90% and specificities of 60–80%, depending on age group and methodology.

• EEG-based tools approved as diagnostic aids (not standalone tests) demonstrate their strongest value in borderline or ambiguous cases.

QEEG is especially informative in:

• adults with unclear childhood history,

• high-functioning individuals with behavioural compensation,

• mixed presentations involving anxiety or mood symptoms.

It adds biological plausibility, not diagnostic finality.

Eye Tracking: Objective Markers of Attentional Control

Eye-tracking technology has emerged as a powerful, non-invasive tool in ADHD assessment.

Research consistently demonstrates that individuals with ADHD show:

• increased saccadic intrusions,

• reduced fixation stability,

• higher rates of gaze variability,

• difficulty maintaining gaze on task-relevant stimuli.

Meta-analytic and experimental studies report:

• classification accuracies ranging from 70–85% when eye-tracking metrics are combined with behavioural tasks,

• particularly strong differentiation during sustained attention and inhibition paradigms.

Eye tracking is valuable because:

• it bypasses self-report entirely,

• it captures micro-level attentional lapses,

• it reflects real-time cognitive control.

It answers a simple but powerful question:

“Does attention drift at the level of perception itself?”

Motion Tracking: Quantifying Hyperactivity and Motor Regulation

Hyperactivity is one of the most inconsistently assessed ADHD symptoms—often judged subjectively.

Motion tracking changes that.

Studies using infrared sensors, actigraphy, or camera-based motion analysis show:

• children and adults with ADHD exhibit significantly higher movement frequency and variability during cognitive tasks,

• motion metrics correlate with symptom severity and functional impairment,

• when combined with CPT or cognitive tasks, motion tracking improves classification accuracy to 75–90% in some experimental models.

Motion tracking is particularly useful in:

• children who suppress hyperactivity during interviews,

• adults with internal restlessness rather than overt hyperactivity,

• cases where observer reports conflict.

It converts “restlessness” from a descriptor into a measurable variable.

Why Serial Multimodal Integration Improves Accuracy

Each tool has limitations. Crucially, their limitations are different.

• Interviews are rich in meaning but vulnerable to bias.

• CPT provides behavioural data but lacks context.

• QEEG adds biology but is probabilistic.

• Eye tracking captures attentional drift.

• Motion tracking quantifies motor regulation.

When integrated serially:

• false positives decrease,

• masked ADHD is uncovered,

• diagnostic confidence rises.

Multimodal models consistently show:

• improvement in diagnostic classification accuracy from ~65–70% (interview alone) to ~80–90% when multiple objective measures converge.

Disagreement between tools does not weaken diagnosis—it forces reconsideration, which is good medicine.

Clinical Impact: Fewer Errors, Better Outcomes

Over-diagnosis leads to unnecessary medication.
Under-diagnosis leads to years of failure and self-blame.

Multimodal assessment reduces both by:

• identifying non-ADHD causes of attention problems,

• clarifying borderline cases,

• guiding personalised treatment planning.

Speed improves because certainty improves, not because corners are cut.

The Direction of Modern ADHD Diagnosis

The future of psychiatric diagnosis is not “tests replacing clinicians.”
It is clinicians augmented by converging data.

Serial integration of:

• clinical interview,

• CPT,

• QEEG,

• eye tracking,

• motion tracking

represents a shift from opinion-based diagnosis to evidence-weighted clinical judgment.

The Take-Home Message

ADHD diagnosis is strongest when:

• meaning, behaviour, biology, and movement are assessed together,

• data is interpreted by a single clinician,

• and conclusions are reached through convergence, not convenience.

This is not complexity for its own sake.
It is precision through integration.

Dr. Srinivas Rajkumar T, MD (AIIMS), DNB, MBA (BITS Pilani)
Consultant Psychiatrist & Neurofeedback Specialist
Mind & Memory Clinic, Apollo Clinic Velachery (Opp. Phoenix Mall), Chennai
✉ srinivasaiims@gmail.com 📞 +91-8595155808