Occupational Cognitive Ecology and Internalizing Symptom Profiles: A LatentVariable Analysis of Backend and NonBackend Developers within a Multidimensional Mental Health Framework

Prof. Dr. Leenah  Ãskaree; Aqsa  Yaqoob; Dr.Mahwish  Saeed; Engineer Ãmmaar  Baig; Ahmad Shujāã  Baig

doi:10.63056/ACAD.003.02.0888

Authors

Prof. Dr. Leenah Ãskaree Chairperson, Department of Psychology, Faculty of Social Sciences and Humanities, Hamdard University Main Campus, Karachi, Pakistan. Post-Doctoral Fellowship at International Islamic University, International Research Institute, Islamabad, Pakistan Author
Aqsa Yaqoob PhD (Scholar), Department of Psychology, University of Karachi, Pakistan Author
Dr.Mahwish Saeed Assistant Professor, Head Psychology programs, Head of Placement, Alumni and Corporate Liaison. Faculty of Allied Health Sciences, Iqra University , North Campus Author
Engineer Ãmmaar Baig Business Analyst, The Resource Group, Karachi, Pakistan Author
Ahmad Shujāã Baig HR Consultant & Student of MPhil Psychology, Department of Psychology, University of Karachi, Pakistan Author

DOI:

https://doi.org/10.63056/ACAD.003.02.0888

Keywords:

anxiety , depression, Multigroup SEM, latentvariable modelling

Abstract

Grounded in the paradigm of occupational cognitive ecology, this study modelled the latent structure linking cognitive load, mental fatigue, and internalizing symptomatology (anxiety, depression) in backend and nonbackend software developers within a multidimensional mental health framework. A crosssectional dataset (N = 320) was analysed via structural equation modelling (SEM), incorporating observed indicators from the Cognitive Load Questionnaire (CLQ), Mental Fatigue Scale (MFS), and standardised anxiety and depression measures, alongside sociodemographic moderators (educational level, family system structure).

The measurement model demonstrated excellent fit (χ²/df = 1.94, CFI = .962, TLI = .955, RMSEA = .054, SRMR = .041), supporting the distinct but correlated latent factors of cognitive load, fatigue, and internalizing symptoms. In the structural model, cognitive load significantly predicted mental fatigue (β = .69, p < .001), anxiety (β = .54, p < .001), and depression (β = .49, p < .001), consistent with cognitiveenergetics theory (Boksem & Tops, 2008). Multigroup SEM revealed that backend developers exhibited steeper loadtofatigue and loadtoanxiety paths than nonbackend developers, reflecting taskspecific cognitive demands.

Educational level moderated the load–fatigue (β = –.23, p < .001) and load–anxiety (β = –.21, p < .001) links, with postgraduates showing attenuated slopes, aligning with metacognitive resource accounts (Efklides, 2008; Eysenck et al., 2007). Family system exerted significant buffering on stress–depression (β = –.11, p < .001) and load–anxiety (β = –.10, p = .002) associations, in line with socioecological resilience perspectives (Triandis, 1995; Uchino, 2009). A threeway interaction (CLQ × Education × Family System; β = –.09, p = .005) confirmed multiplicative protective effects, with postgraduates in extended families demonstrating the lowest predicted fatigue under high load.

These findings advance the field by integrating latentvariable modelling with occupational mentalhealth theory, highlighting how rolespecific cognitive demands, educational resources, and familial support interact to shape internalizing symptom profiles. Implications include the need for precisiontargeted interventions that align loadmanagement strategies with the cognitive ecology of specific developer roles.