ALI
Alexithymia Language Index
ALI helps therapists detect and track alexithymia-related language patterns across sessions, so they can target interventions and monitor change over time.
ALI is a product concept and research direction.
Alexithymia: Hidden in plain sight
Why is Alexithymia often missed?
Alexithymia is frequently missed or misdiagnosed because its symptoms can be mistaken for other conditions (depression, autism), and many healthcare providers aren't trained to recognize it, leading people to go undiagnosed their whole lives.
Often missed
Top reasons it gets overlooked
- Mostly internal: there aren’t consistent outward signs.
- Easy to misread: it can look like depression, neurodivergence traits, or “stress.”
- No shared label: people may not know there’s a name for it.
- It’s a trait: not an illness, so it’s often overlooked—even by the person experiencing it.
Symptoms
What are the common symptoms of alexithymia?
- Difficulty identifying feelings (DIF)
- Difficulty describing feelings (DDF)
- Externally oriented thinking (EOT)
- Low emotion vocabulary (e.g., “fine”, “off”, “stressed”)
- Going blank when asked about feelings
- Emotional flatness, numbness, or shutdown
- Feeling overwhelmed but unable to name it
- Irritability without clear emotional meaning
- Physical distress without clear cause
- Body cues not linked to emotions
- Quick problem-solving when emotion rises
- Events and logistics emphasized over inner meaning
- Difficulty expressing needs and boundaries
- Conflict escalation or shutdown
- Attachment insecurity / relationship strain
- Delayed clarity (“I realize later what I felt”)
Why did we build ALI?
ALI automatically scans session language for auditable evidence of alexithymia-related patterns. ALI can detect alexithymia-related patterns in session language and link them to excerpts.
What this changes
Instead of relying on memory or a single screening moment, ALI continuously looks for language patterns that support or contradict alexithymia facets.
ALI is not a diagnostic tool. It’s designed to support clinical reflection with transcript-linked evidence.
Making Alexithymia Visible
- Reduce shame: name the pattern without blame.
- Improve care: choose clearer, earlier interventions.
- Track change: see whether emotional clarity is improving over time.
ALI at a glance
A quick overview of what ALI measures and how it is used in therapy.
ALI charts
These are the newest chart designs from the dashboard client summary, shown here with example data.
What the three ALI facets mean
These facets describe three common ways people get stuck with feelings. Click each one to see what it looks like in conversation and what therapists often focus on next.
Facet preview
Difficulty Identifying Feelings
A person feels something, but cannot tell what it is.
Often sounds like
- “I do not know what I am feeling.”
- “Something is off, but I cannot tell what.”
- “I feel numb.”
Intervention
Slow down, notice body cues, and offer a short menu of possible feelings.
Alexithymia Scales:
TAS vs PAQ vs ALI
A quick, high-level comparison of the widely used self-report scales VS. our conversation-derived ALI approach.
TAS-20
Toronto Alexithymia Scale
- 20-item self-report checklist.
- Global alexithymia score plus three subscales.
- Widely used benchmark in research and clinics.
PAQ-24
Perth Alexithymia Questionnaire
- Self-report with positive and negative emotion facets.
- Richer profile of identifying and describing feelings.
- Still relies on people rating themselves on items.
ALI
Alexithymia Language Index (Feelpath)
- Conversation-derived indices from session language.
- Facet scores tied to concrete, labeled excerpts.
- Designed to complement TAS/PAQ, not replace them.
More charts (legacy)
Earlier explorations of additional ways to visualize alexithymia-related patterns.
Bars in the yellow or red bands suggest stronger alexithymia signals for that composite.
Difficulty Appraising Feelings
Negative vs. positive emotion.
Identifying vs. Describing
Noticing a feeling vs. putting it into words.
Index: 45 (within typical range).
After session 1: Baseline + pattern
A clear starting point: severity, where the client gets stuck, and a few working hypotheses to test.
Overall severity (G-DAF)
68 / 100
possible alexithymia range
Strongest facet
G-DDF 78
difficulty describing feelings stands out
EOT (G-EOT)
45
more balanced attention on inner vs outer experience
What feels clinically useful
They can report distress, but struggle to name and describe what they feel, especially around negative affect. Start with gentle labeling and body-to-feeling links.
This can later switch from example numbers to transcript-derived indices, while keeping the same story.
Generic feeling terms
18 → 9
“fine”, “bad”, “stressed”, “tired”
Specific emotion words
6 → 22
more differentiated labels over time
“I don’t know what I feel” phrases
7 → 2
less uncertainty language
Somatic-only mentions
10 → 4
body cues increasingly linked to feelings/needs
In a future version, each bar could link to supporting excerpts (for example, moments of shutdown during criticism).
Top “moves” that helped
- Affect labeling prompts (“If you had to pick one: sad, angry, scared?”) → more specific emotion words.
- Emotion wheel / menu → faster differentiation (anger vs hurt vs fear) when they are unsure.
- Body cue → feeling → need → stronger “because” statements and clearer next steps.
What often got in the way
- Shame about having feelings, especially in interpersonal topics.
- Switching quickly into events and logistics when emotion rises.
- Using broad labels (“stressed”) without narrowing to a primary emotion.
Start of session
60-second body check-in: “Where do you feel it in your body?” Then offer 3 candidate emotions and ask which is closest.
When they go blank
Slow down and normalize: “This makes sense.” Then scaffold: body cue → emotion family → meaning/need → small action.
This section is meant to reduce uncertainty and support clinical judgment, not replace it.