@inproceedings{feltman-etal-2026-discriminant,
title = "Discriminant Validity: Disentangling Health and Emotional Constructs from Language-Based Assessments",
author = "Feltman, Scott and
V Ganesan, Adithya and
Ringwald, Whitney and
Schwartz, H. Andrew and
Kotov, Roman and
Luft, Benjamin and
Boyd, Ryan and
Kjell, Oscar",
editor = "Zirikly, Aya and
Bar, Kfir and
MacAvaney, Sean and
Ireland, Molly and
Ophir, Yaakov and
Atzil-Slonim, Dana and
Varadarajan, Vasudha and
Bedrick, Steven and
Desmet, Bart",
booktitle = "Proceedings of the 10th Workshop on Computational Linguistics and Clinical Psychology ({CLP}sych 2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.clpsych-1.4/",
pages = "43--55",
ISBN = "979-8-89176-421-7",
abstract = "Language-based assessments have demonstrated high convergent validity with corresponding mental and physical health constructs, however often fail to address discriminant validity - the measure{'}s ability to distinguish the target construct from related ones. This is a common phenomenon within the domain of mental health, as well as comorbidity with physical health conditions. Identifying key features of individual dimensions of mental and physical health present in language can unlock new avenues of research for natural language processing and psychology. We propose two augmentations to the objective function of the Ridge model, deriving closed-form solutions compatible with Singular Value Decomposition-based solvers, to enforce discriminant validity of off-target constructs using Mean Squared Error (MSE) and Squared Cosine Similarity (SCS,) both having widespread use in contrastive learning. By varying the discrimination strength, we find that a decrease in 0.005 Pearson correlation points can result in a Pearson correlation point increase upwards of 0.132 in discriminant validity for mental and physical health constructs derived from self-reported questionnaires. We see similar improvements across multiple fundamental psychopathology dimensions simultaneously, increasing discriminant validity by 0.012 with stronger increases coming from more noisy, less reliable constructs. Our contributions provide a theoretically grounded path towards improving confidence in language-based assessments in the clinical sector, improving specificity of said assessments to various areas of health."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="feltman-etal-2026-discriminant">
<titleInfo>
<title>Discriminant Validity: Disentangling Health and Emotional Constructs from Language-Based Assessments</title>
</titleInfo>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="family">Feltman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adithya</namePart>
<namePart type="family">V Ganesan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Whitney</namePart>
<namePart type="family">Ringwald</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">H</namePart>
<namePart type="given">Andrew</namePart>
<namePart type="family">Schwartz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roman</namePart>
<namePart type="family">Kotov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Luft</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="family">Boyd</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Oscar</namePart>
<namePart type="family">Kjell</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 10th Workshop on Computational Linguistics and Clinical Psychology (CLPsych 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aya</namePart>
<namePart type="family">Zirikly</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kfir</namePart>
<namePart type="family">Bar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sean</namePart>
<namePart type="family">MacAvaney</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Molly</namePart>
<namePart type="family">Ireland</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaakov</namePart>
<namePart type="family">Ophir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dana</namePart>
<namePart type="family">Atzil-Slonim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vasudha</namePart>
<namePart type="family">Varadarajan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Bedrick</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bart</namePart>
<namePart type="family">Desmet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-421-7</identifier>
</relatedItem>
<abstract>Language-based assessments have demonstrated high convergent validity with corresponding mental and physical health constructs, however often fail to address discriminant validity - the measure’s ability to distinguish the target construct from related ones. This is a common phenomenon within the domain of mental health, as well as comorbidity with physical health conditions. Identifying key features of individual dimensions of mental and physical health present in language can unlock new avenues of research for natural language processing and psychology. We propose two augmentations to the objective function of the Ridge model, deriving closed-form solutions compatible with Singular Value Decomposition-based solvers, to enforce discriminant validity of off-target constructs using Mean Squared Error (MSE) and Squared Cosine Similarity (SCS,) both having widespread use in contrastive learning. By varying the discrimination strength, we find that a decrease in 0.005 Pearson correlation points can result in a Pearson correlation point increase upwards of 0.132 in discriminant validity for mental and physical health constructs derived from self-reported questionnaires. We see similar improvements across multiple fundamental psychopathology dimensions simultaneously, increasing discriminant validity by 0.012 with stronger increases coming from more noisy, less reliable constructs. Our contributions provide a theoretically grounded path towards improving confidence in language-based assessments in the clinical sector, improving specificity of said assessments to various areas of health.</abstract>
<identifier type="citekey">feltman-etal-2026-discriminant</identifier>
<location>
<url>https://aclanthology.org/2026.clpsych-1.4/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>43</start>
<end>55</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Discriminant Validity: Disentangling Health and Emotional Constructs from Language-Based Assessments
%A Feltman, Scott
%A V Ganesan, Adithya
%A Ringwald, Whitney
%A Schwartz, H. Andrew
%A Kotov, Roman
%A Luft, Benjamin
%A Boyd, Ryan
%A Kjell, Oscar
%Y Zirikly, Aya
%Y Bar, Kfir
%Y MacAvaney, Sean
%Y Ireland, Molly
%Y Ophir, Yaakov
%Y Atzil-Slonim, Dana
%Y Varadarajan, Vasudha
%Y Bedrick, Steven
%Y Desmet, Bart
%S Proceedings of the 10th Workshop on Computational Linguistics and Clinical Psychology (CLPsych 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-421-7
%F feltman-etal-2026-discriminant
%X Language-based assessments have demonstrated high convergent validity with corresponding mental and physical health constructs, however often fail to address discriminant validity - the measure’s ability to distinguish the target construct from related ones. This is a common phenomenon within the domain of mental health, as well as comorbidity with physical health conditions. Identifying key features of individual dimensions of mental and physical health present in language can unlock new avenues of research for natural language processing and psychology. We propose two augmentations to the objective function of the Ridge model, deriving closed-form solutions compatible with Singular Value Decomposition-based solvers, to enforce discriminant validity of off-target constructs using Mean Squared Error (MSE) and Squared Cosine Similarity (SCS,) both having widespread use in contrastive learning. By varying the discrimination strength, we find that a decrease in 0.005 Pearson correlation points can result in a Pearson correlation point increase upwards of 0.132 in discriminant validity for mental and physical health constructs derived from self-reported questionnaires. We see similar improvements across multiple fundamental psychopathology dimensions simultaneously, increasing discriminant validity by 0.012 with stronger increases coming from more noisy, less reliable constructs. Our contributions provide a theoretically grounded path towards improving confidence in language-based assessments in the clinical sector, improving specificity of said assessments to various areas of health.
%U https://aclanthology.org/2026.clpsych-1.4/
%P 43-55
Markdown (Informal)
[Discriminant Validity: Disentangling Health and Emotional Constructs from Language-Based Assessments](https://aclanthology.org/2026.clpsych-1.4/) (Feltman et al., CLPsych 2026)
ACL
- Scott Feltman, Adithya V Ganesan, Whitney Ringwald, H. Andrew Schwartz, Roman Kotov, Benjamin Luft, Ryan Boyd, and Oscar Kjell. 2026. Discriminant Validity: Disentangling Health and Emotional Constructs from Language-Based Assessments. In Proceedings of the 10th Workshop on Computational Linguistics and Clinical Psychology (CLPsych 2026), pages 43–55, San Diego, California, USA. Association for Computational Linguistics.