"Phone in the Room, Mind on the Roam": Investigating the Impact of Mobile Phone Presence on Distraction

This study explores whether the passive presence of a mobile phone affects attentional control among young adults (18–25), addressing a gap distinct from active phone use research. Given ubiquitous smartphone exposure and concerns about cognitive costs, the authors investigate how mere presence may drain cognitive resources (brain drain) or trigger automatic attention, potentially impairing performance. The work is motivated by mixed findings in prior literature and the lack of evidence in the Greek context. The study formulates five hypotheses: H1: passive phone presence impairs attentional task performance; H2: higher self-reported distraction correlates with poorer attention; H3: higher fear of losing the phone (nomophobia) exacerbates distraction when the phone is absent; H4: higher smartphone addiction associates with poorer attentional performance; H5: women report higher nomophobia than men.

Mobile phones are pervasive and can interrupt attentional processes through notifications, salience, and relevance to users’ goals, potentially recruiting automatic attention and reducing available capacity (Kahneman; Ward et al.). The mere presence effect posits cognitive costs even without active use, with some studies reporting decrements in working memory and fluid intelligence (Ward et al., 2017) and sustained attention (Thornton et al., 2014), while others find null or context-dependent effects (Hartmann et al., 2020; Skowronek et al., 2023). Notifications and phone-related cues capture attention involuntarily (Stothart et al., 2015; Roye et al., 2007). Mobile phone addiction involves excessive, compulsive use linked to attentional lapses and poorer well-being; university students are particularly vulnerable (Kwon et al.; Long et al.). Nomophobia—the fear of being without a phone—relates to anxiety and obsessive-compulsive tendencies and is prevalent among students, with some evidence for gender differences (SecurEnvoy, 2012; Yildirim & Correia, 2015). Despite conceptual overlap, nomophobia and addiction are distinct but interrelated patterns with debated nosology. Overall, findings on smartphone presence, addiction, and nomophobia show complex, sometimes contradictory relationships with attention.

Design: Between-groups experimental study with random assignment to mobile phone presence (experimental) vs. absence (control). Participants: N = 144 young adults (75 females, 69 males), ages 18–25, mean age = 21.5 (SD = 3.0). Recruitment via snowball sampling; inclusion required phone ownership. Sample size planning used G*Power (d = 0.5, α = 0.05, power = 0.80), minimum n = 128; final n exceeded requirement. Measures: (1) Smartphone Distraction Scale (SDS; 16 items; subscales: attention impulsiveness, online vigilance, multitasking, emotion regulation). Response format adapted to 6-point Likert; reliability acceptable (α 0.74–0.88; CFA: CFI = 0.91, TLI = 0.90, RMSEA = 0.07). (2) Smartphone Addiction Scale–Short Version (SAS-SV; 10 items; total 10–60; cut-offs: >31 males, >33 females; α = 0.83). (3) Nomophobia Questionnaire (NMP-Q; 20 items; α = 0.94). All instruments translated and back-translated (English–Greek–English). ANT: The Attention Network Test assessed alerting, orienting, and executive control via reaction times and error rates, with cue conditions (no, central, double, spatial) and flanker types (congruent, incongruent, neutral). Trials: fixation 400–1600 ms; cue 100 ms; target up to 1700 ms; responses via left/right keys. Procedure: Data collection Jan–Mar 2024 in a controlled lab environment (temperature, noise, lighting). Participants tested individually; alternation-based randomization to conditions. Control group left phones outside; experimental group placed phones on desk in view, data/Wi-Fi on, sound/vibrate enabled; instructed not to touch phones during ANT. Informed consent obtained; debriefed post-task. Analysis: IBM SPSS v27. Three multivariate outliers removed via Mahalanobis distance, final n = 141 reflected in descriptive table. Computed descriptive stats, Cronbach’s α, t-tests (one-tailed for H1), Pearson correlations, and two-way ANOVA (addiction level × phone presence) for ANT errors and time.

Baseline comparisons showed no significant group differences in age, sex, and phone engagement. Performance: ANT time did not differ between groups (Absent: n = 71, M = 23,628.8 s, SD = 6181.0; Present: n = 70, M = 22,783.1 s, SD = 5808.8), t(139) = 0.837, p = 0.202, d = 0.14. ANT mistakes did not differ (Absent: M = 2.7, SD = 3.8; Present: M = 2.0, SD = 2.8), t(139) = 1.248, p = 0.107, d = 0.21. Correlations: SDS, NMP-Q, and SAS-SV were strongly interrelated (r = 0.66, 0.78, and 0.67 respectively; all p < 0.001). Under phone presence, ANT mistakes correlated with SDS Attention Impulsiveness (r = 0.25, p < 0.05) and SAS-SV (r = 0.37, p < 0.01), indicating more errors with higher impulsiveness/addiction. NMP-Q showed a non-significant relation to errors in phone presence (r = 0.20). Under phone presence, task time had small negative, non-significant correlations with SDS Multitasking and SAS-SV (both r = −0.15). ANOVA: Addiction level had a main effect on ANT errors, F(1,137) = 4.548, p = 0.035, η²p = 0.032; no significant interaction with phone presence; no significant effects for ANT time. Gender differences: Women reported higher nomophobia than men, t(139) = 6.175, p < 0.001, d = 1.04; means 78.5 vs. 59.8 (difference = 18.7 points). Collectively, H1 not supported; H2 broadly not supported (except weak error associations under phone presence); H3 not supported; H4 partially supported (errors only); H5 supported.

Contrary to expectations and some prior studies on mere presence effects, passive smartphone presence did not impair attentional performance on the ANT. The authors suggest habituation to ubiquitous smartphones and task demands may attenuate distraction, and that methodological differences across studies (active use vs. mere presence, notification salience, task complexity) contribute to mixed findings. The lack of correlation between self-reported distraction and objective performance implies a perception–performance gap potentially shaped by metacognitive control or task load. Smartphone addiction showed a small yet significant effect on errors, especially when phones were present, indicating that individual differences in maladaptive use may moderate susceptibility to distraction. Nomophobia related to subjective distraction but not performance, suggesting anxiety-related vigilance may not translate to measurable attentional deficits in this context. Robust gender differences in nomophobia reaffirm demographic influences. Overall, results indicate that cognitive impacts of smartphone presence are nuanced, context-dependent, and moderated by individual differences rather than universal.

Among young adults, the passive presence of a smartphone did not significantly degrade attentional performance on the ANT, although higher smartphone addiction related modestly to increased errors, and women reported substantially higher nomophobia than men. These findings highlight the importance of individual differences and task context over mere device presence. Future work should employ longitudinal and ecologically valid designs, manipulate task difficulty and phone salience, and examine moderating variables (e.g., personality, coping, cognitive load) to refine theory and inform targeted interventions and policies for managing digital distractions in educational and occupational settings.

Key limitations include: snowball sampling and potential selection bias; a sample size adequate for planned analyses but potentially limited for broader generalizability; omission of control for individual difference variables (e.g., baseline cognitive abilities, fatigue, habitual smartphone usage patterns); measurement constraints—ANT may not capture all attention facets relevant to smartphone distraction; adaptation of scales to a six-point Likert format without independent validation or explicit permission from original authors; reliance on a controlled lab setting that may not reflect real-world environments; removal of multivariate outliers may affect representativeness. Future studies should use larger, more diverse samples, include additional cognitive and behavioral measures, validate any instrument modifications, and test in real-world contexts.