Abstract
<p>This paper introduces Algorithmic Trait Amplification (ATA), a theoretical construct describing the process by which social media recommendation algorithms may selectively detect, reinforce, and progressively intensify pre-existing narcissistic personality traits in users through iterative exposure to trait-congruent content. Situated within the broader conceptual framework of the Digital Narcissus Effect, introduced here as a structural analogy to Ovid's myth of Narcissus and Echo, ATA is proposed as a five-stage cyclical model in which subclinical narcissistic traits generate preferential engagement with narcissism-congruent content; algorithmic systems capture this engagement as a preference signal; curated content environments progressively mirror and validate narcissistic self-perception; corrective psychological experiences are constrained; and narcissistic trait expression may be intensified along the subclinical continuum over time. Importantly, ATA does not necessarily imply stable trait restructuring in all users, but may involve intensified trait expression, reinforcement of maladaptive self-schemas, or increased behavioural manifestation of narcissistic tendencies under conditions of sustained algorithmic exposure. Drawing on personality psychology, cyberpsychology, algorithmic studies, clinical theory, and developmental psychology, this paper argues that ATA may represent one significant mechanism, among several interacting contextual factors, contributing to observed trends in narcissistic trait expression in algorithmically saturated social environments. The construct offers a potential explanatory framework relevant to ongoing and contested debates about population-level changes in narcissistic trait distribution and identifies the recommendation algorithm as a potentially anti-therapeutic environment for individuals with narcissistic predispositions. Implications for clinical practice, platform regulation, digital literacy education, and future empirical research are discussed, along with the methodological challenges such research will need to address.</p>