Skip to main content
SearchLoginLogin or Signup

Smartphones Hinder Prospective Memory in Users With High Levels of Smartphone Dependency

Volume 4, Issue 3. DOI: 10.1037/tmb0000113

Published onAug 10, 2023
Smartphones Hinder Prospective Memory in Users With High Levels of Smartphone Dependency


Prospective memory (PM) has rarely been investigated in the context of smartphones. We embedded a PM task in an online, smartphone-based survey that participants were asked to take on their smartphone, and we investigated PM performance in relation to self-report on strategies used, smartphone distractions, and smartphone dependency measures. Of the 478 participants who accepted the survey job through the Amazon Mechanical Turk platform, 295 total participants were included in the study as they met full inclusionary criteria, passed the PM attention check cue, and had complete data. Our sample was aged 21–71 years old (M = 38.03; SD = 10.90) and was 62.5% male. Measures included a PM task that required participants to respond “N/A” to a question presented later in the survey and two questionnaires on smartphone dependency. The true purpose of the study was not disclosed in the PM portion of the survey to avoid ceiling effects, which are common in PM research. After completion of this portion and debriefing of participants, we asked what PM strategies they had used (e.g., attributing importance to the task and using reminders) and whether smartphone-based and self-initiated distractions had occurred during the survey. Overall, a third of the participants were successful on the smartphone PM task. We found that higher likelihood of PM success was predicted by (a) higher self-reported importance of the task; (b) lower use of external reminders; and (c) lower levels of smartphone dependency. Higher likelihood of PM success was also associated with fewer reported smartphone- and self-initiated distractions. This association was not influenced by levels of smartphone dependency. Findings suggest that smartphones are a hindrance to PM in those with significant smartphone dependency and in those who engage in smartphone-related distractions. Smartphones may induce competing social motives, possibly rendering some traditional PM strategies inefficient. Future research could decrease this social competition through mindfulness applications: In our sample, insight into use proved to be a potential protective factor against the negative effects that smartphone dependency and distractions have on PM performance.

Video Abstract

Keywords: prospective memory, smartphones, smartphone dependency, nomophobia, smartphone distractions

Funding: This research received no external funding.

Disclosures: The authors have no known conflicts of interest to disclose.

Author contributions: Holly Elizabeth Phelps contributed in conceptualization, data curation, formal analysis, investigation, methodology, and writing–original draft. Claudia Jacova contributed in conceptualization, methodology, supervision, and writing–review and editing.

Data availability: There have been no prior uses of these data. Data and study materials are publicly available to other researchers at

Open Science Disclosures: The data are available at

Correspondence concerning this article should be addressed to Claudia Jacova, School of Graduate Psychology, Pacific University, 222 Southeast 8th Avenue, Hillsboro, OR 97123, United States. Email: [email protected]

Have you ever opened your smartphone to do a task but got distracted by a notification, application, or other phone-related obligation, only to forget your original intention? This common occurrence would be classified as a prospective memory (PM) failure in the context of smartphone use. PM—or remembering future intentions (Ellis, 1996)—is a topic of interest in the field of cognition due to its importance in everyday life. Because PM oftentimes involves other people, its failure affects not only our own lives but also our reputation and relationship with others. A number of research studies have investigated social context factors that influence prospective remembering (Brandimonte et al., 2010; Cicogna & Nigro, 1998; Kobayashi & Maruno, 1994; Kvavilashvili, 1987; Schaefer & Laing, 2000). These researchers have found that a prosocial motive—telling participants that remembering an intention is important to another person—resulted in better PM performance than either a control condition or a personal gain motive in the form of a course credit (Brandimonte et al., 2010). However, when the prosocial motive was presented in combination with an external reward (i.e., personal gain motive), this facilitating effect disappeared (Brandimonte et al., 2010). Thus, prosocial or altruistic motives have a positive impact on prospective remembering, but an added personal gain motive can neutralize this impact. Such social competition effects on PM should be considered when operating smartphones—our ubiquitous, trusty companions and portals into endless social opportunity. Constant social competition and reward could somehow affect PM intentions and strategies one has when using a smartphone, though this relationship is unknown.

Smartphones provide a complex social context. They are used to satisfy social needs (Li & Chung, 2006; Veissière & Stendel, 2018), but these may range from maintaining connection with others to avoiding awkward face-to-face interactions (Dwyer et al., 2018; Kim, 2017; Kushlev, Dwyer, & Dunn, 2019; Kushlev, Hunter, et al., 2019; Rotondi et al., 2017; Sbarra et al., 2019). There are likely implications of smartphone-specific social motives for prospective memory, but these are not the focus of the two common theoretical perspectives on smartphone technology and its impact on human memory to date. One perspective focuses on the utility of smartphones as mnemonic aids. Early research found that smartphones can serve as compensatory mechanisms for PM difficulties and provide benefits above and beyond using written reminders (e.g., setting alarms; Bos et al., 2017; Evald, 2015). PM research on postbrain injury has shown significant task improvement in these individuals when smartphone reminders were provided, which led to additional benefits of increasing feelings of independence, confidence in coping with memory deficits, and general mood (Ferguson et al., 2015). External reminders in general have been deemed more ecologically valid while providing meaningful improvement in PM, though these studies tend to include small samples (Jones et al., 2021). Smartphone-based memory training has also led to increased working memory performance in older adults, though subjective memory contentment can remain unaffected (Oh et al., 2018). Greater usage of smartphone features to assist daily remembering in older adults has also been found to improve management of activities of daily living regardless of age, base cognition, and initial smartphone acumen (Scullin et al., 2022). The other perspective considers smartphones as potential sources of interference with memory processes. Such research has documented the deleterious effects that smartphone and technology use can have on long-term recall (Frein et al., 2013; Henkel, 2014; Sparrow et al., 2011), working memory (Cain et al., 2016; Uncapher et al., 2015), and memory for spatial construction (Burnett & Lee, 2005; Parush et al., 2007). One recent study by Hartmann et al. (2020) assessed the effect of the physical presence of smartphones on different memory functions, with the aim of extending previous findings of smartphone presence having negative effects on cognitive performance (i.e., “brain drain”; Thornton et al., 2014; Ward et al., 2017). They assessed the effect of a smartphone’s physical presence on a PM task administered during a working memory task on a laptop. They found that smartphone presence did not have an overall effect on PM but interacted with smartphone dependency, such that individuals with low smartphone dependency performed better when the smartphone was absent whereas those with high dependency did not show this advantage. Additionally, PM performance decreased with increasing smartphone dependency, regardless of smartphone presence. Of note, dependency relationships only existed for one of the measures used in the study.

With the growing ubiquity of smartphones in everyday life, dependency behaviors have become a prominent phenomenon (Park, 2019; Park & Kaye, 2019). About 46% of U.S. smartphone owners consider their smartphone as something they cannot live without (Smith, 2015). The compulsive nature of smartphone use combined with the frequency of use across contexts is illustrative of smartphone dependency (Chen et al., 2017; Lee et al., 2014; van Deursen et al., 2015). Park (2019) described two types of smartphone dependency: functional dependence and existential dependence. Functional dependence involves purposeful use and utilization of the smartphone (e.g., using a calculator function to determine a server’s tip; Park, 2019). Existential smartphone dependence, rather, is contingent upon use that promotes extensions of the self or connection with others—satisfying a need for attachment, gaining access to others, avoiding isolation, rejection, or even disconnection (Park, 2019). These social components connect closely with the term “nomophobia” or “no mobile phone phobia,” that is, the fear of being without one’s mobile device (Bhattacharya et al., 2019; King et al., 2013). Nomophobia can result from device overuse and is largely related to the inability to maintain connectivity with social networks (Nikhita et al., 2015). Nomophobia has the potential to develop in both functional and existential use.

The study by Hartmann et al. (2020) indicates that smartphone presence and smartphone dependency jointly impact PM. However, it should be noted that their design did not involve direct use of the smartphone during the PM task. We reasoned that active use of the smartphone during the PM task could reveal additional effects of smartphone dependency, and designed our task with this reasoning in mind. Not only are users exposed to smartphone distractions, but users even actively seek distractions while using the device. The effects of these smartphone-based distractions (both coming from the phone and user initiated) on PM performance are unknown, but it is likely that the demands of the PM task must compete with the levels of one’s smartphone dependence and susceptibility to smartphone distractions. This combination can be perceived as one’s level of social connectedness to their device. Therefore, the purpose of our study was to shed light on PM performance when it involves the use of a smartphone. To do so, we first investigated the strategies participants used in our novel online PM task in order to determine whether these are comparable to those established in previous research, and that our task truly measures PM proper (see Graf & Uttl, 2001). This allowed us to determine the convergent validity of our task. Next, we investigated the pattern of relationships between smartphone dependency, smartphone distractions, and PM performance. We used two smartphone dependency measures to probe functional and existential dimensions of dependency in relation to PM performance. We were also interested in how smartphone-related distractions affected PM performance, since both place demands on attentional resources. Prior research on interruptions/distractions when executing a PM task has proven to be detrimental both in laboratory and naturalistic settings (Dismukes, 2012; McDaniel et al., 2004). We distinguished between internal (self-initiated) and external (smartphone delivered) smartphone distractions based on current evidence that different sources of distraction may produce different cognitive costs (Unsworth & McMillan, 2014). Furthermore, previous work has shown that the attentional cost of mere smartphone presence is greater for those with higher smartphone dependency (Ward et al., 2017). We therefore examined whether smartphone dependency moderated the effect of distractions on PM.

Our study had three aims. The first aim was to identify the specific PM strategies that were used. We hypothesized that PM success was associated with four strategies: use of external reminders, internal monitoring, occurrence of recollections, and attribution of importance. The second aim was to examine the relationship between smartphone dependency and PM. We hypothesized that PM success was related to low scores on measures of nomophobia and smartphone involvement. The third aim was to determine the relationship between PM success and smartphone distractions and to evaluate the role of smartphone dependency in this relationship. We hypothesized that PM success was predicted by low numbers of smartphone-based and self-initiated distractions. We also hypothesized that this relationship was moderated by smartphone dependency: We expected that the adverse effect of distractions on PM was more evident for individuals with high rather than low dependency scores.


This study sought approval by the Pacific University institutional review board. All participants gave electronic informed consent at the outset of the survey. As compensation, participants were given $2 for their participation and completion of the survey. Funding for participant participation was provided by research lab funds at the university. Procedures were executed in conformity with the approved guidelines. All study materials are available online at Jacova and Phelps (2023).

Because ceiling effects may occur when participants have insight into the purpose of PM tasks, which can be a byproduct of remembering processes becoming attentional rather than mnemonic (i.e., vigilance rather than episodic), we used deception as part of the study design. Participants were initially not told about the true nature of the experiment (Lezak et al., 2012; Uttl, 2008; Wilson et al., 2008), but rather that we were interested in understanding smartphone usage patterns, smartphone dependency behaviors, and overall smartphone habits. Debriefing and reasoning about the true purpose of the experiment occurred after completion of the smartphone usage questionnaire, where we asked whether participants had guessed or known the true purpose of the survey. Apart from the study purpose, all other study details and descriptions remained factual (i.e., inclusion criteria, potential risks and benefits, and reward/compensation opportunities) with the exception of an added risk due to the initial deception.

Participants and Design

A total of 478 participants accepted the survey job after being recruited through the Amazon Mechanical Turk (MTurk) platform. To be included in the study, participants were required to be 18 years or older, be able to read and understand written and spoken instructions and survey questions in English, be geographically located within the United States, be verified MTurk workers. In addition, they were asked to take the survey on a smartphone. The participants had to select all of these inclusionary statements in order to proceed to taking the survey. If they did not select all statements including using a smartphone for the survey, they were immediately directed to the end of the survey (participants were not informed that they must select all statements in order to proceed). A total of 95 participants did not meet inclusionary criteria, and of these, 78 individuals indicated that they were not taking the survey on a smartphone. An additional 54 participants did not pass the prospective memory attention check question at the beginning of the survey (see next section) and were thus excluded from the study. Last, 34 additional participants had incomplete data (e.g., left the study before finishing) and were excluded from the study. No further exclusionary criteria were applied, and technological acumen was assumed because of participants’ status as MTurk workers. After excluding these participants, we had a total of 295 participants with complete data. The sample was aged 21–71 years (M = 38.03; SD = 10.90); 62.5% male.

The study design was correlational and consisted of administering a novel PM task within an online smartphone-based survey that participants were asked to take on their smartphones. This was to ensure that PM performance was impacted by smartphone use in tandem with relevant, individual differences (i.e., smartphone dependency). We chose the online format because of the restrictions on in-person research that began in March 2020 as part of COVID-19 social distancing precautions.


PM Task

Our task was event-based and designed to measure PM proper (see Graf & Uttl, 2001). We used a nonfocal task to minimize the potential for ceiling effects. This type of task requires more effortful processing to notice the cue than is typical for focal tasks where an obvious discrepancy exists between the cue and ongoing task (Ball et al., 2019; Brewer et al., 2010; Einstein & McDaniel, 2005). The task components were embedded in the survey, with instructions given at the beginning as part of orienting participants to the survey after the informed consent and inclusion criteria. An attention check, commonly used in MTurk jobs, was presented to verify that participants read the instructions. Instructions stated:“There was an error made in the creation of this survey. When you get to the question about the kind of smartphone plan you have, please choose “N/A”. To indicate that you have read and understood these instructions, please select the second box below before clicking Next. ”If participants did not select the second box before clicking next, their data were excluded. PM performance was classified as successful when N/A was chosen as the answer to the critical item, or as unsuccessful if any other response was chosen. It should be noted that the other answer options (individual or shared smartphone plans) were assumed to capture the smartphone plans of all participants. Therefore, N/A was only a plausible response option if prospective remembering occurred.

Smartphone Dependency Surveys

To assess levels of smartphone dependency, two questionnaires were incorporated within the study survey after the PM instructions: the Nomophobia–Questionnaire (NMP-Q; Yildirim & Correia, 2015) and the Mobile Phone Involvement Questionnaire (MPIQ; Walsh et al., 2010). These two surveys were used in a recent study conducted by Fryman and Romine (2020) examining the consequences of nomophobia and its comorbidities.

The NMP-Q is a 20-item, self-report nomophobia questionnaire. All 20 items are rated using a 7-point Likert scale from 1 = strongly disagree to 7 = strongly agree. Scores from each item are summed up to obtain a total score ranging from 20 to 140, with higher scores corresponding to greater nomophobia severity. Range interpretations of total scores include scores of 20 = absence of nomophobia; scores > 20 but < 60 = mild levels of nomophobia; scores > 60 but < 100 = moderate levels of nomophobia, and scores > 100 = severe levels of nomophobia (min = 20, max = 140). The NMP-Q was the first questionnaire developed and validated to measure nomophobia among U.S. college students. Yildirim and Correia (2015) found a four-factor structure using exploratory technique: not being able to communicate, losing connectedness, not being able to access information, and giving up convenience. The NMP-Q was shown to produce valid and reliable scores. Internal consistency was excellent for the total score (Cronbach’s α = .939) and good for each of the four individual dimensions (α = .939, .847, .827, and .814, respectively).

The MPIQ is an eight items, self-report measure of mobile phone involvement based broadly on components of behavioral addiction (Brown, 1993, 1997; Walsh et al., 2010). Each item gauges an individual component: cognitive and behavioral salience, interpersonal conflict, conflict with other activities, euphoria, loss of control, withdrawal, relapse, and reinstatement. Items are rated using a 7-point Likert scale from 1 = strongly disagree to 7 = strongly agree, with higher scores indicating higher involvement (min = 8, max = 56). A principal component analysis revealed that all eight items were assessing a unitary construct, explaining 39.62% of the variance. Items were then summed and averaged with a reliability analysis to reveal that the MPIQ had moderate reliability (α = .78).

Smartphone Usage Questions

The usage questions formed the “ongoing task” and were not included in the analyses. Items were adapted from various questionnaires (e.g., Front Porch Communities and Services, 2016; Marketest, 2019) to acquire information about various smartphone use patterns and history (e.g., length of smartphone ownership; frequency of use, application use, data plan) as standard questionnaires about smartphone use do not currently exist. These questions varied in response format to include multiple choice, open response, and yes/no questions. The length of this ongoing task was approximately 15 min, designed to allow for short-term memory to be cleared of the PM cue to ensure that the PM proper task did not become a task of PM vigilance (>15 min according to Shapiro & Krishnan, 1999).

Follow-Up Questions

Twenty-seven follow-up questions were asked to gather further information about the participants’ overall experience throughout the PM task. The following strategies were examined: external reminders (four items), monitoring and checking (four items), recollection effects (two items), and importance effects (one item). For three of the predictor variables—external reminders, internal monitoring, and recollections—items were summated. Table 1 shows a summary of questions for each strategy and associated Cronbach’s α. For importance effects, only one question was asked: How important was it for you to remember to answer “N/A” for the question asking about smartphone plans?

Table 1

Questions Associated With PM Strategies Used


External reminders

Internal monitoring



Did you write a note to remember to choose “N/A” for the question asking about smartphone plans?

Did you, at random points, mentally check in with yourself to remember to choose “N/A” for the question asking about smartphone plans?

Did you at any point randomly recall to remember to choose “N/A” for the question asking about smartphone plans?

Did you use a strategically placed object (e.g., moving a paperweight or object in sight, putting a rubber band on your wrist, or changing something else about your environment) to serve as a reminder to choose “N/A” for the question asking about smartphone plans?

How often did you randomly think about answering “N/A” for the question asking about smartphone plans?

How often did you randomly think about answering “N/A” for the question asking about smartphone plans?

Did you set a reminder on your smartphone or use your smartphone in any way to remember to choose “N/A” for the question asking about smartphone plans?

Did you strategically check in with yourself to remember to choose “N/A” for the question asking about smartphone plans?

Did you use any other reminders to remember to choose “N/A” for the question asking about smartphone plans?

Did you start to check in with yourself more as you approached the question asking about smartphone plans?

Cronbach’s α




Note. PM= prospective memory.

Questions about aspects of internal and external distractions prior to the PM cue were also asked. For self-implemented distractions, we asked one question:“Did you spontaneously back out of the survey at any point before answering the question asking about smartphone plans to engage in another task on your phone not prompted by a notification? (e.g., visiting a social media app, checking emails, checking a bank statement, etc.). ”For distractions emanating from the smartphone, we focused on smartphone notification distractions that could plausibly have arisen. These are listed in Table 2. We summed these four items for analysis.

Table 2

Questions Associated With Smartphone Distractions


  1. During the survey, did you receive a phone call before answering the question asking about smartphone plans?

  1. During the survey, did you receive a text message before answering the question asking about smartphone plans?

  1. During the survey, did you receive a battery warning message before answering the question asking about smartphone plans?

  1. During the survey, did you receive any other notification before answering the question asking about smartphone plans? (e.g., Facebook message, bank account message, appointment notification, phone update notification, an alarm going off)

Cronbach’s α = .899


Age, sex, race, ethnicity, and education were considered as potential covariates. These demographic characteristics were collected in order to monitor for between-group differences in PM success and to control for such factors in further analyses.

Statistical Analyses

Binary logistic regression was used to examine the associations between continuous variables and the binary PM outcome (odds of success vs. failure). Chi-square analyses were used to examine the relationships between categorical variables and PM performance. We adjusted all regression models for age because of the wide age range of our participant sample, and due to the possibility that age-related variability in our variables of interest might bias the findings. Additionally, we investigated the relationship of sex, race, ethnicity, and education with PM performance, but only included these as covariates if statistically significant associations emerged. We used logistic regression to evaluate the relationship between the PM task and PM strategies used during the survey. We also used logistic regression to investigate the associations between smartphone dependency, smartphone/self-implemented distractions, and PM performance. Last, binary logistic regression interaction models were used to examine the relationship between distractions (smartphone-based or self-initiated) and PM performance as moderated by smartphone dependency. Due to conducting multiple regression analyses, significance values were set at .01 to reduce the likelihood of Type I error (Tabachnick & Fidell, 2007).


Descriptive Data

Overall, a little over one third of participants (36.6%) were successful on the smartphone PM task (N = 108). In terms of PM strategies, we found that 24.41% of participants reported the use of external reminders, 62.71% reported internal monitoring, 65.76% reported recollections, and 88.47% considered the PM task at least somewhat important. NMP-Q total scores had a wide range and reasonably normal distribution, with two thirds of the sample reporting moderate or severe nomophobia. Our data showed internal consistencies comparable to those originally reported: α = .967 for the total NMP-Q, α = .960 for “not being able to communicate,” α = .926 for “losing connectedness,” α = .849 for “not being able to access information,” and α = .858 for “giving up convenience.” The MPIQ total score also had a wide range and a somewhat platykurtic distribution. Internal consistency for our data was α = .913 for the total MPIQ. Both smartphone-based and self-initiated distractions were endorsed by a small percentage of participants (16.3% and 23.1%, respectively). Descriptive data on smartphone dependency, self-implemented, and smartphone-based distractions are reported in Table 3.

Table 3

Descriptive Data



M (SD)



% of PM success in group

Skewness (SE)

Kurtosis (SE)

Age (range = 50)


38.03 (10.90)


.886 (.142)

.143 (.283)

Q1 (21–34 years)




Q2 (35–48 years)




Q3 (49–62 years)




Q4 (63–71 years)
































 High school education or less




 Some college




 College degree




NMP-Q total (range = 120)


87.78 (29.77)


−.491 (.142)

−.718 (.283)

 No Nomoph. (range ≤ 20)




 Mild Nomoph. (range = 21–60)




 Mod. Nomoph. (range = 61–100)




 Severe Nomoph. (range > 100+)




NMP-Q subscales

 Not being able to access information (range = 24)


19.79 (5.73)


−.751 (.142)

−.021 (.283)

 Giving up convenience (range = 30)


22.36 (7.54)


−.453 (.142)

−.620 (.283)

 Not being able to communicate (range = 36)


26.56 (10.50)


−.550 (.142)

−.779 (.283)

 Losing connectedness (range = 30)


19.07 (8.92)


−.100 (.142)

−1.259 (.283)

MPIQ total (range = 48)


33.63 (11.69)


−.201 (.142)

−1.032 (.283)

SP distractions sum


 Zero distractions




 One distraction




 Two distractions




 Three distractions




 More than four distractions




Self-implemented SP distractions










Note. The gender demographic variable had an N = 293 due to omitting two individuals who did not identify as neither male nor female. All other demographic subsamples had an N = 295. Q = quartile; PM = prospective memory; Nomophob. = nomophobia; SP = smartphone; MPIQ = Mobile Phone Involvement Questionnaire; SE = standard error; NMP-Q = Nomophobia–Questionnaire.

Demographic Factors and PM Success

Table 3 shows the demographics characteristics for our sample and the percentage of PM success for each demographic group. Age was not associated with PM performance at the chosen p level (≤.01). The unstandardized coefficient (years) was B = .023, SE = .011, Wald = 4.367, p = .037; (OR = 1.023, 95% CI [1.001, 1.046]); however, our model showed that with increasing age, there was greater likelihood of PM success. Racial status (White or Non-White), ethnicity (Hispanic or non-Hispanic), education (high school or less, some college, or a college degree), and gender (male vs. female) were not predictive of PM performance. Based on these findings, no demographic variables except age were included in subsequent models.

PM Performance and Strategies Used (Aim 1)

Table 4 shows the results of the binomial regressions for all predictors and all specified interactions, as well as the percentage of PM success versus failure where applicable. Overall percentages of PM success included the use of the following strategies: 4.07% used external reminders, 19.32% used internal monitoring strategies, 23.05% reported recollections, and 34.92% considered the task at least somewhat important to remember. Only external reminders, model fit = χ2(2) = 32.849, p < .001; Nagelkerke R2 = .144, and self-reported level of importance of remembering the PM cue, model fit = χ2(2) = 40.929, p < .001; Nagelkerke R2 = .177, were related to likelihood of PM success. A higher number of external reminders used was associated with a lower likelihood of PM success. In fact, 11.11% of participants with PM success and 32.09% of participants with PM failure used external reminders. All types of external reminders that we measured were significantly associated with the lower likelihood of PM success: strategically placed objects, χ2(1, 295) = 17.17, p < .001; note reminders, χ2(1, 295) = 13.54, p < .001; other external reminders, for example, Google assistant, message, timer, etc., χ2(1, 295) = 12.90, p < .001; and smartphone-based reminders, χ2(1, 295) = 10.19, p < .001.

Table 4

Factors Relating to Likelihood of PM Success and Associated Performance Within Both Success and Failure Groups





Wald χ2






PM+ % or M (SD)

PM− % or M (SD)

External remindersa




















Internal monitoring





































































74.58 (28.60)

95.19 (27.84)



















27.88 (9.84)

36.72 (11.32)










Smartphone distractionsa




















Self-implemented SP distractionsa




















SP Distractions × SP Dependency (NMP-Q)


















SP Distractions × SP Dependency (MPIQ)


















S-I Distractions × SP Dependency (NMP-Q)


















S-I Distractions × SP Dependency (MPIQ)


















On the other hand, higher ratings of self-reported importance were related to greater likelihood of PM success, which is reflected by the high percentage ratings of importance in this group (95.37%) compared to the PM failure group. Internal monitoring strategies, model fit = χ2(2) = 7.141, p = .028; Nagelkerke R2 = .033, and recollections, model fit = χ2(2) = 2.122, p = .035; Nagelkerke R2 = .031, were not related to likelihood of PM success at the selected p level. It is important to note that the models for internal monitoring strategies and recollections did not show to be a good fit for the data.

PM Performance and Smartphone Dependency (Aim 2)

Table 4 shows that both level of nomophobia, model fit = χ2(2) = 38.292, p < .001; Nagelkerke R2 = .166, and level of mobile phone involvement, model fit = χ2(2) = 47.410, p < .001; Nagelkerke R2 = .203, were related to likelihood of PM success. Higher ratings of both were related to lesser likelihood of PM success (see Table 4). Post hoc analyses were conducted to determine the subcategories of nomophobia (NMP-Q) related to PM failure. All subcategories were analyzed within the same model. Out of these four categories (not being able to access information, giving up convenience, not being able to communicate, and losing connectedness), only “losing connectedness” was significantly related to the likelihood of PM success within the model. Those endorsing greater fear of losing connectedness had decreased likelihood of PM success, B = −.155, SE = .029, Wald = 29.444, p < .001; (OR = .857, 95% CI [.810, .906]); model fit χ2(5) = 67.688, p < .001; Nagelkerke R2 = .280. Post hoc analyses were also conducted on subcategories of mobile phone involvement (MPIQ), where each of the eight questions measures a distinct aspect of smartphone behavior (cognitive salience, behavioral salience, interpersonal conflict, conflict with other activities, euphoria, loss of control, withdrawal, and relapse and reinstatement). Of these eight areas, behavioral salience (“I often use my mobile phone for no particular reason”) and interpersonal conflict (“arguments have arisen with others because of my mobile phone use”) were significantly related to the likelihood of PM success: Those who admitted to often using their smartphone were more likely to have PM success, B = .291, SE = .108, Wald = 7.303, p = .007; (OR = 1.338, 95% CI [1.803, 1.653]), and those who had more arguments were less likely to have PM success, B = −.540, SE = .121, Wald = 19.810, p < .001; (OR = .583, 95% CI [.460, .739]), model fit χ2(9) = 97.912, p < .001; Nagelkerke R2 = .386.

PM Performance and Smartphone Distractions (Aim 3)

Distractions were classified as either smartphone distractions (i.e., distractions coming from the smartphone in the form of notifications or other messages/alerts) or self-implemented distractions (i.e., the user backing out of the survey to autonomously engage in a separate task on the smartphone). Overall, of those who engaged in smartphone distractions, 14.58% of participants had PM success and 85.42% participants had PM failure. Of those who engaged in self-implemented smartphone distractions, 2.94% of participants had PM success and 97.06% participants had PM failure. Table 4 shows that number of smartphone distractions, model fit = χ2(2) = 26.883, p < .001; Nagelkerke R2 = .119, and self-implemented smartphone distractions, model fit = χ2(2) = 65.737, p < .001; Nagelkerke R2 = .273, were significantly related to likelihood of PM success. Higher levels of both smartphone distractions and self-implemented smartphone distractions were related to lower likelihood of PM success. When looking at specific types of smartphone distractions, none were significantly related to the likelihood of PM success. Therefore, it is only the sum of these distractions that is significantly related to the decreased likelihood of PM success.

We also examined smartphone dependency as a moderator of the relationship between distractions and PM success (four separate models). As shown in Table 4, none of the interaction terms were statistically significant, but, in each of the four models, the measure of smartphone dependency was the only significant predictor of PM success. Specifically, level of nomophobia (NMP-Q) was the only statistically significant predictor when smartphone distractions and their interaction with the NMP-Q were in the model, B = −.021, SE = .005, Wald = 20.547, p < .001; (OR = .979, 95% CI [.970 .988]). Mobile phone involvement (MPIQ) was the only significant predictor of PM success when smartphone distractions and their interaction with MIPQ were in the model, B = −.062, SE = .012, Wald = 24.442, p < .001; (OR = .940, 95% CI [.918, .963]). Nomophobia alone predicted PM success when self-implemented smartphone distractions and their interaction with nomophobia were in the model, B = −.016, SE = .005, Wald = 11.777, p < .001; (OR = .984, 95% CI [.975, .993]). Finally, mobile phone involvement alone predicted PM success when self-implemented smartphone distractions and their interaction with mobile phone involvement were in the model, B = −.037, SE = .013, Wald = 8.804, p < .001; (OR = .964, 95% CI [.940, .989]).


The present study is the first naturalistic experimental design using deception to investigate PM performance while using the smartphone and to examine its relationship with smartphone dependency. A PM task specifically designed for the present study was administered in the course of an online smartphone-based survey. We found a positive relationship between self-reported task importance and PM success but no or inconsistent relationships emerged between other strategies and PM success. Successful performance on the PM task was associated with lower levels of smartphone dependency on the measures of nomophobia and mobile phone involvement. Finally, successful PM performance was associated with fewer smartphone distractions (both self- and smartphone-initiated). However, contrary to our expectations, these associations were not moderated by measures of smartphone dependency.

PM Strategies and Task Validity

Our finding that higher self-reported level of importance predicted greater likelihood of PM success aligns with most previous research on importance effects (Einstein et al., 2005; Kvavilashvili, 1987; Meacham & Singer, 1977; Penningroth & Scott, 2013; Walter & Meier, 2017). The use of external reminders had a negative relationship with likelihood of PM success, which is contrary to what has been previously reported (e.g., Dismukes, R. K et al., 1990; Jones et al., 2021; McDaniel et al., 2004). One possible explanation for this inverse relationship is that external reminder users also gave low importance ratings regarding the PM task and thus may have designated insufficient or weak cues (Penningroth & Scott, 2013). Another possibility is that if participants set external reminders on their smartphones, this unfortunately could have been a distraction in and of itself and could have possibly been acting as an interfering, external reward, as was found in the research by Brandimonte et al. (2010). In other words, a potential social competition component could be at play, deeming external reminders when operating the smartphone less effective. Though external reminders in general have been associated with meaningful improvement in PM, this research has also typically included small samples and has not focused on the hindering aspect of smartphone use on external reminders (Jones et al., 2021); therefore, results of our study may shed light on what once was considered an effective method of remembering future intentions, being less reliable and more hindering in the context of operating a smartphone.

Possible reasons for these null findings include the questionable efficacy of these strategies in the context of smartphone use, a suboptimal approach in measuring these strategies (e.g., see Harris, 1980), and not accounting for the differentiation between recollections (i.e., remembering exact details) and familiarity (i.e., feelings of knowing but lacking specific detail). We conclude that our task is a valid measure of PM in the context of smartphone use, given our convergent findings of relationships that would be expected based upon previous research; however, we acknowledge that the specific demands of prospective remembering in the context of smartphones are unknown and may differ from the demands of standard PM tasks. For this reason, individuals may not rely on the traditional strategies for remembering intentions.

PM and Smartphone Dependency

Our study revealed that with increasing levels of smartphone dependency, participants had a decreased likelihood of PM success. We measured smartphone dependency using two established measures: NMP-Q and MPIQ. Of the four subcategories of the NMP-Q, increasing levels of “losing connectedness” was significantly related to the decreased likelihood of PM success. This finding aligns with previous research on nomophobia and device overuse (Nikhita et al., 2015). A similar inverse relationship with PM was also found by Hartmann et al. (2020) on one of their dependency measures (Ward’s scale but not Smartphone Addiction Scale). Our findings draw attention to methodological differences: In our study participants performed the PM task on their smartphones, hence the effects might be more marked than when the smartphone is simply physically present on the desk as was the case in the Hartmann study. Our findings also raise the possibility that some, but not other facets of smartphone dependency, have negative implications for PM. We found that on the NMP-Q, increasing levels of “losing connectedness” and, on the MPIQ, more interpersonal conflict (“arguments have arisen with others because of my mobile phone use”) were related to decreased PM success. Unfortunately, Hartmann et al. did not analyze subscales or specific domains on their smartphone dependency measures, but we could speculate that the relationship between dependency and PM may be specific to social and interpersonal facets of dependency.

The finding of a positive relationship between behavioral salience and PM success may seem counterintuitive at first glance, but the awareness involved in being able to identify one’s overactive smartphone use may indicate increased insight into use behavior or dependency. Awareness of dependency could be a key mechanism to counteract smartphone dependency. For example, recent research has found that higher self-control in older adults has been associated with lower prevalence of problematic smartphone usage (Busch et al., 2021), a finding that aligns with our study results: Higher smartphone dependency is associated with decreased likelihood of PM success, yet insight of one’s smartphone use could be a protective factor against PM failures. Furthermore, Busch et al. (2021) found this balance between awareness–dependency relationship to occur despite one of the top reasons for smartphone use in their older adult sample being social media use (i.e., social connectedness)—a type of use that has been linked to worse everyday memory functioning in adulthood (Frein et al., 2013; Soares & Storm, 2018; Tamir et al., 2018). Increased social competition could act as a strong distraction operating in the background of memory tasks, leading to decreased likelihood of PM success. Yet, older adults with high levels of awareness are somehow resistant to such distraction. This is interesting given that traditionally, greater processing demands of background tasks have rendered older adult PM performance worse than that of their younger peers, particularly when task execution must be delayed (Craik, 1986; Craik & Jennings, 1992; Einstein et al., 2000; Salthouse, 1991). The results of our delayed-execute PM task of increasing age trending toward higher likelihood of PM success may point to this background task phenomenon being different in the context of smartphone use for older adults, potentially through a type of resiliency stemming from insight into use. At the very least, awareness could enable responses to regulate dependency and reduce dependency burden.

To summarize, we found a pattern of both negative and positive associations with PM in relation to smartphone use. This pattern suggests different components of dependency have different implications for PM. It is possible that the distinction between functional and existential dependency holds relevance in interpreting this difference. It has been noted in previous work that functional dependency is characterized by a degree of awareness and a willingness to change dependent behaviors (Park, 2019). Such dependency may therefore not be an immovable obstacle to PM success. By contrast, existential dependency has been described as an ontological extension of self, with the smartphone becoming an important part of a person’s identity. Individuals with existential dependence have been found to have less insight into their smartphone use, as they do not readily admit a dependence problem, nor do they realize how heavily they depend upon their smartphone in daily life despite the technological usefulness (Park, 2019). It is conceivable that existential dependency may result in reduced attention to the needs of others, including failing to follow up on plans and future intentions.

Smartphone Distractions

Our study found that increases in both smartphone and self-implemented smartphone distractions were related to the decreased likelihood of PM success. These findings align with previous research on the detriment of interruptions on PM in both lab-based and naturalistic settings (Dismukes, 2012; D(Dismukes et al., 2007); McDaniel, et al., 2004). Furthermore, both distraction types took their toll on PM performance regardless of level of smartphone dependency. In all interaction models, smartphone dependency measures were uniquely and strongly predictive of PM performance. We can speculate that completing the survey and PM task on the smartphone itself may have involved competing social motives, pitting distractions against the PM task. Specifically, because external rewards can negatively interfere with social motives if existing in tandem (Brandimonte et al., 2010), smartphone distractions may operate like external rewards due to underlying motivations for engagement. These reward-based distractions in turn can interfere with social motives both in and outside the context of smartphone use, negatively affecting the social facilitation of PM success. Our findings extend the “brain drain” research of Ward et al. (2017) by implicating motivational factors in the cognitive effects of technology. Unlike Ward et al., we failed to find a moderating role for smartphone dependency in the adverse effects that distractions had on PM performance. It is likely that differences in task design (smartphone used for task vs. smartphone merely present) account also for this discrepancy.


Our data collection occurred during the COVID-19 pandemic restrictions, which may have impacted the social connectedness needs of our sample. It is possible that had our study been conducted during a different time, our sample demographics and overall results may have revealed a different picture. Due to the online format, we had no way of confirming that participants were taking the survey on a smartphone device. When examining the research on patterns of MTurk smartphone users, one study found that MTurk workers were more likely to use smartphones for completing jobs later in the evening (Casey et al., 2017). Five of our eight survey batches released spanned more than 1 day to completion, indicating that tasks could have been fulfilled at many points in time over multiple days. Additionally, Jacques and Kristensson (2017) found that though only 7% of MTurk workers exclusively use a mobile device to access, MTurk jobs, workers showed high rates of readiness to switch from computers to mobile devices when requested. Specifically, 82.7% of MTurk workers accepted tasks from their desktop computer and switched to their mobile device to continue the task (Jacques & Kristensson, 2017).

Concerns have arisen in the literature about MTurk workers failing to represent some segments of the U.S. population (Pew Research Center, 2016), thus impacting the generalizability of our study. Studies have found that in many respects, MTurk participants mostly reflect the U.S. population as a whole with regard to racial and ethnic background (with the exception of African Americans), income distribution (under $150,000), and gender identity (slightly more female 57%; Moss et al., 2020). The majority of our sample was White, non-Hispanic, and more represented by males (62.5%). Age distribution of MTurk workers was the least representative of the U.S. population, and this was also reflected in our truncated sample’s age range. Research indicates a large amount of MTurk workers are between the ages of 18 and 39 (66.5% compared with United States 31.6%). Likewise, the two largest quartiles in our MTurk sample were those ages 21–34 and 35–48. The most drastic difference in age occurs in the older adult population (6% compared with United States 35.6%). Our sample also showed that older adults (ages 63–71) were less represented (4.7%). Currently, no research exists on how the cognitive abilities of MTurk workers compare to the U.S. population.

A key design feature of our study was instructing participants to take our survey on their smartphones; however, we had no supplementary data to confirm that participants took the survey on their smartphone other than participants initially endorsing an inclusionary criterion to confirm this use. Older adults may have been underrepresented in our sample because they chose not to participate in our study given that it involved the use of a smartphone. However, we note that the underrepresentation of older adults in our sample aligns with age demographics typically reported amongst MTurk workers. Nonetheless, our ability to generalize the results with the population of older adults is limited. Alternatively, older adults may have decided to use a computer or laptop, which may have facilitated their PM performance. Covarying for age, however, did not change our findings. Additionally, recent trends of increased technology use amongst older adults have been reported. Specifically, according to a 2022 study by Pew Research Center, smartphone ownership in older adults rose from 13% in 2012 to 61% in 2021 (Faverio, 2022). From 2018 to 2021, the increase in smartphone ownership for adults age 65+ was the largest amongst any other age group. Similar trends were conveyed in a 2020 report by the American Association for Retired Persons (Kakulla, 2020) finding that 51% of their 2,607 older adult sample had bought some technology product within the last year, with the top purchases being smartphones (23%). According to Zhu and Cheng (2022), while the digital divide is objectively inevitable in relation smartphone use, there is significant diversity among older adults, with many not conforming to technology-averse stereotypes.

Another limitation is our use of mild deception to mask our focus on PM. It is possible that participants guessing or knowing the true purpose of the survey were more likely to have PM success than those being unaware of it. Based on responses to the follow-up question, the deception worked for over half of participants (57.8%), who reported having no idea at all that the study was about PM. These participants were also more likely to have PM success. Therefore, it appears unlikely that awareness contributed to PM success. It is also possible that the disclosure of this deception impacted responses to the final portion of questions about PM strategies and distractions. A final limitation was that we simply covaried for age. Our study was not designed to investigate age-related differences in the effects we identified (we had uneven numbers across age groups and small numbers in the older groups). We acknowledge that such differences are likely to exist and should be systematically investigated in future research.

Implications and Future Directions

Our results extend the findings of previous research on the brain drain hypothesis. Our findings suggest that smartphones are a hindrance to PM functioning in individuals with significant smartphone dependency, and that smartphone distractions are essential to this hindrance. We also found that external reminders may not be as effective in the context of smartphone use, particularly when there is a low level of importance in remembering the future intention. Higher importance, however, was associated with PM success. Future research could further investigate how external reminders (as well as internal monitoring) could be helpful in the context of smartphone use, perhaps by investigating how to increase the level of importance of a future intention to increase chances of PM success. Additionally, it is possible that smartphones induce social motives that compete with the natural social environment. Future research should also examine the hypothesis of competing social motives in appropriate experimental studies. To decrease social competition, future intervention research could investigate the utility of applications to enhance the awareness of smartphone use leveraging mindfulness-based principles. Such app exercises, if successful, could help decrease the negative effects that smartphone dependency has on cognitive and social functioning. This could better ensure that users are operating their devices in a socially responsible manner.

Supplemental Materials

Received February 3, 2023
Revision received May 8, 2023
Accepted May 30, 2023
Rayon Ston:

If you are a student seeking top-notch academic assistance, look no further than our "Assignment helper". Wе undеrstand thе challеngеs studеnts facе in mееting dеadlinеs and maintaining acadеmic еxcеllеncе. Our tеam of еxpеriеncеd professionals is dеdicatеd to providing comprеhеnsivе support tailorеd to your spеcific nееds. Our sеrvicеs еncompass a widе rangе of subjеcts and disciplinеs, еnsuring that we can assist you with any assignmеnt. From еssays and rеsеarch papеrs to complеx projects, wе'vе got you covеrеd. Our commitmеnt to quality, affordability, and timеly dеlivеry sеts us apart. With our onlinе assignmеnt hеlp in thе UK, you can rеliеvе thе strеss of academic dеmands and focus on your studiеs. Wе'rе hеrе to еmpowеr your academic journey and hеlp you achiеvе your goals. Explorе our sеrvicеs today to еxpеriеncе acadеmic еxcеllеncе likе nеvеr bеforе!