Printer Friendly

Using Smartphones While Walking is Associated with Delay but Not Social Discounting.

Delay discounting refers to the finding that subjective reward value decreases as the time to receive a given reward increases. The process that underlies delay discounting has been discussed in great detail by researchers in both behavioral economics (e.g., Frederick, Loewenstein, & O'Donoghue, 2002; Loewenstein, Read, & Baumeister, 2003; Vuchinich & Heather, 2003) and psychology (e.g., Green & Myerson, 2004; Madden & Bickel, 2010; Odum, 2011). Delay discounting is typically measured by a task that includes choices or decisions between an immediate and a delayed reward. Although human and nonhuman animals often prefer the immediate reward rather than the delayed reward, even when the reward is larger for the delayed option, the extent to which immediate or delayed rewards are preferred varies depending on individual and environmental factors. If individuals tend to choose an immediate smaller reward compared to a delayed larger one, then their choices are referred to as impulsive. On the other hand, individuals that select delayed larger rewards are considered to engage in more self-controlled choices (Critchfield & Kollins, 2001). When the subjective reward values are plotted as a function of delay, the curves represent the delay discounting function. Although there is debate about what function can best describe the delay discounting slope (e.g., Laibson, 1997; Mazur & Biondi, 2009; McKerchar et al., 2009), recent studies have shown that the following hyperbolic functions formulated by Mazur (1987) provide a good fit to the data.

V = A/(1 + kD) (1)

where V is the subjective value of a delayed reward, A is the amount of the delayed reward, and D is the delay to the receipt of the reward. The parameter k is a constant that represents delay sensitivity and varies among individuals. A larger k means that subjective reward value decreases more rapidly as delay increases.

In addition to providing a novel impulsivity measure, delay-discounting researchers have also demonstrated that discounting is related to socially significant problem behaviors. For example, many addictive behaviors are closely associated with the degree of discounting, such as substance abuse (Coffey, Gudleski, Saladin, & Brady, 2003; Madden, Bickel, & Jacobs, 1999), cigarette smoking (Bickel, Odum, & Madden, 1999; Mitchell, 1999), pathological gambling (Alessi & Petry, 2003; Dixon, Marley, & Jacobs, 2003), and internet addiction (Saville, Gisbert, Kopp, & Telesco, 2010).

In addition, smartphone use, which has become a ubiquitous human behavior, also leads to a steeper discounting rates (Hadar et al., 2017; Tang, Zhang, Yan, & Qu, 2017). Reed, Becirevic, Atchley, Kaplan, and Liese (2016) pointed out that smartphone-based text messaging can also be classified as a form of addictive behavior. Furthermore, even moderate smartphone use in combination with other activities can cause problem behaviors related to carelessness or inattention. For example, texting while driving (TWD; e.g., Caird, Johnston, Willness, Asbridge, & Steel, 2014; Stavrinos et al., 2013) is not only careless, but also dangerous for the person TWD and others around them. Thus, TWD is a violation of road traffic law in many countries such as the United States, where TWD is banned for drivers in 47 states and the District of Columbia (Insurance Institute for Highway Safety, 2018; McCartt, Kidd, & Teoh, 2014).

Hayashi, Russo, and Wirth (2015) measured TWD and examined the relationship between TWD and delay discounting with a hypothetical monetary reward. They found that participants with higher rates of TWD discounted delayed rewards to a greater extent than control participants. Hayashi et al.'s (2015) study indicated that delay discounting is related not only to smartphone use, but also to behaviors that are dangerous for the participant and others around them. In addition to the delay-discounting task with a monetary reward, Hayashi, Miller, Foreman, and Wirth (2016) examined the relationship between TWD and delay discounting using a novel discounting task in which participants rated the possibility of replying immediately to a text message received while driving, versus waiting to reply for a set period of time (see also Atchley & Warden, 2012; Reed et al., 2016). They found that participants with higher TWD rates discounted delayed opportunity to reply to a greater extent, whereas no obvious difference among groups was observed in the rate of discounting with a monetary reward. The latter result is inconsistent with Hayashi et al. (2015). Hayashi et al. (2016) pointed out that this inconsistency may be due to procedural differences between two studies. Hayashi and Blessington (2018) recently extended research on distracted behavior by texting to the classroom (i.e., texting in the classroom) by examining the relationship between texting in the classroom and delay discounting using two delay-discounting tasks similar to Hayashi et al. (2016). They found that the participants who showed higher frequencies of texting in the classroom also exhibited steeper discounting functions for both tasks.

As with TWD, USWW or distracted walking can also be a dangerous behavior. USWW includes activities while walking such as texting a message like TWD, playing a game, watching a video, etc. USWW is responsible for many traffic accidents, just like TWD (see Mwakalonge, Siuhi, & White, 2015 for review). It is not surprising that as smartphone use has increased, so has the prevalence of pedestrian injuries related to smartphone use (Nasar & Troyer, 2013). In fact, the number of injuries for pedestrians exceeded that for drivers in 2010 (1,506 for pedestrians vs. 1,162 for drivers). In Japan, USWW has also emerged as a social problem. According to a survey by the Tokyo Fire Department (2018), 152 people were taken to the hospital by ambulance due to accidents caused by USWW and while riding a bicycle between 2010 and 2014 within their jurisdiction. As a result, several cities (e.g., Fort Lee, New Jersey and Honolulu, Hawaii in the United States) have created legislation to discourage USWW with penalties such as fines (Gates, 2013; Mohn, 2017; Parks, 2017). The statistic of "injury due to USWW" shown in these population statistics is supplemented by laboratory studies demonstrating reduced performance in different interactive tasks (Haga et al., 2015; Schwebel et al., 2012). Thus, participants engaging in USWW have a higher risk of accidents, similar to TWD. Taken together, we predict that USWW is also correlated with delay discounting. However, there is currently no study that addresses this relationship, which is the first aim of this study.

Social discounting, which is another type of discounting, may also be related to distracted behaviors such as TWD and USWW. The concept of social discounting was first proposed by Jones and Rachlin (2006), who indicated that the amount of money that a participant is willing to forgo in order to share a certain amount of money with someone else declines as a function of social distance between the participant and the person with whom they are sharing the money. Similar to delay discounting, the decrease in the amount of money to be kept can be described by a hyperbolic function as follows:

v = V/(1 + sN) (2)

where v is the discounted reward value, Vis the undiscounted reward value at social distance = 0, and N is the social distance. The parameter 5 is a constant that represents social distance sensitivity and differs among individuals: the greater the s, the steeper the social discounting curve. Just as the delay discounting task measures aspects of impulsivity, this social discounting task measures the degree to which individuals discount their generosity across social distances. A larger 5 denotes less generosity (i.e., increased selfishness), whereas a smaller's indicates more generosity.

Recent studies have illustrated several factors that affect social discounting. For example, Rachlin and Jones (2008) reported that social discounting rates are higher for larger magnitude rewards than smaller rewards, which is called a "reverse magnitude effect." In addition, various other factors have been studied, such as cultural differences (Romanowich & Igaki, 2017; Strombach et al., 2014), nonadult participants (Sharp et al., 2012), sexual differences (Olson, Rosso, Demers, Divatia, & Killgore, 2016), cigarette smoking (Bradstreet et al., 2012), licit drug use (Romanowich & Igaki, 2017), real and hypothetical rewards (Locey, Jones, & Rachlin, 2011), anonymity (Locey & Rachlin, 2015), and risk (Jin, Pei, & Ma, 2017). Distracted behavior may be associated with social discounting, because TWD and USWW can be dangerous to others and therefore can be thought of as selfish.

Several studies have indicated a potential correlation between delay and social discounting. For example, Rachlin (2002) and Rachlin and Locey (2011) discussed the possibility that one type of discounting could be derived from the other type of discounting, or vice versa. Furthermore, Rachlin and Jones (2008) showed a significant positive correlation between social discounting and delay discounting. If both types of discounting are correlated with each other, regardless of whether one can be derived from the other or a common underlying variable produces both types of discounting, one might expect that the same tendency can be obtained when both types of discounting are compared directly among participants (e.g., Myerson, Green, Hanson, Holt, & Estle, 2003) or between groups (e.g., Holt, Green, & Myerson, 2003). In terms of distracted behavior, if those participants who often indicate distracted behavior show steep delay discounting, then they should also show steep social discounting.

However, other studies have not always obtained a correlation between delay and social discounting and a common third variable. For example, Jones and Rachlin (2009) examined the correlation between the amount of the donation in a public goods game (third variable) and the degree of delay and social discounting. Although they confirmed that steepness of delay and social discounting correlated with each other within participants, they did not observe a clear correlation between steepness of delay discounting functions and the amount donated. The steepness of the social discounting function significantly correlated with the donation amount. Bradstreet et al. (2012) examined the performance of pregnant smokers and nonsmokers on delay and social discounting tasks, and found that pregnant smokers exhibited steeper social discounting functions than nonsmokers, but that no correlation existed between the degree of delay discounting and smoking habit. Wainwright, Green, and Romanowich (2018) showed a significant positive relationship between delay discounting and participant's body mass index (BMI), but no clear association between social discounting and BMI. Although careful assessment is required to determine whether social and delay discounting are independent processes, or whether both types of discounting are related to each other, just as distinct discounting processes may exist for delay and probability discounting, respectively (see Green & Myerson, 2004 for review), delay and social discounting may also reflect separate processes. If this is the case, it is possible that both discounting processes are separately correlated with distracted behaviors, or like smoking, donations, and BMI, one is correlated with distracted behaviors whereas the other is not.

In this study, we focused on the relationship between USWW and delay discounting. By examining the association with another type of distracted behavior (i.e., USWW), this study contributes to the literature that has shown the relationship between delay discounting and a variety of problem behaviors related to carelessness. In addition, we examined the relationship between USWW and social discounting to 1) determine whether one type of distracted behavior that is considered dangerous to others is correlated with selfishness, as measured by social discounting, and 2) determine whether delay and social discounting were correlated with each other.



Two-hundred thirty-nine undergraduate students (74 females and 165 males) at a Japanese university voluntarily participated in the experiment. Participants were at least 18 years old and enrolled in the first author's introductory psychology class. Participants completed a demographic survey during the class, two written discounting tasks, and questions about USWW. Prior to the experiment, participants were provided an explanation about how participants' rights and privacy were protected, and provided informed consent. The explanation and informed consent were presented in writing and orally to all participants. All participants received course credit for their participation. The university's Review Board approved this study.


After giving informed consent, participants were given a questionnaire that consisted of instructions for completing the study, demographic questions, discounting tasks, and questions relating to USWW. First, participants completed four demographic questions including self-reported nationality, gender, college year, and age. Next, participants completed the delay and social discounting task using a paper-and-pencil format. The order of discounting tasks was counterbalanced across participants. After completing the discounting tasks, participants completed questions concerning USWW. Participants were instructed not to flip over and read the following pages unless they had completed the task on that given page. Total time for completing the study was approximately 30 minutes.

Delay Discounting

The delay-discounting task began with an instruction page, followed by seven successive delay-value conditions assigned to their own page. During the delay discounting task, participants were asked to make choices between delayed and immediate hypothetical outcomes. The instructions were as follows:
   The following experiment asks you to make some decisions
   between monetary alternatives. On the next seven
   pages you will be asked to make a series of decisions
   based on your preferences. On each line, you will be
   asked if you would prefer to receive an amount of money
   delivered after a delay (A) and an amount delivered
   immediately (B). Please circle A or B for each line.

Delay values were as follows: 1 week, 2 weeks, 1 month, 6 months, 1 year, 5 years, and 25 years. Delay value order was counterbalanced across participants. At the top of each delayvalue condition page, the following instructions were shown (Below is an example for the 6-month delay):
   Now imagine the following choices between an amount
   of money delivered after 6 months (A) and an amount
   delivered immediately (B). Circle A or B to indicate
   which you would choose in each line.

On each of seven delay values, participants made 10 choices between a delayed and immediate outcome. The delayed option was 100,000 [yen] available after the fixed delay and remained unchanged across all choices (US$1 was equivalent to about 106 [yen] at the time of the study). The immediate option was an amount of money available immediately, which started at 100,000 [yen] and decreased by yen 10,000 to 10,000 [yen] within each delay value. On each page the 10 choices were presented in columns, with the left-hand column containing the delayed option, and right-hand column containing the immediate options.

Social Discounting

The social discounting task also started with an instruction page, followed by seven social-distance conditions, each assigned to their own page. The social discounting task originally developed by Jones and Rachlin (2006) was used in this study. The instructions asked participants to imagine making a list of 100 people in accordance with social proximity. In the list, the closest person to the participant would be at position #1, whereas a casual acquaintance would be #100. The instructions were as follows:
   The following experiment asks you to imagine that you
   have made a list of the 100 people closest to you in the
   world ranging from your dearest friend or relative at
   position #1 to a mere acquaintance at #100. The person
   at #1 would be someone you know well and is your
   closest friend or relative. The person at #100 might be
   someone you recognize and encounter but perhaps you
   may not even know their name. You do not have to
   physically create the list--just imagine that you have
   done so.

   On the next seven pages you will be asked to make a
   series of decisions based on your preferences. On each
   line, you will be asked if you would prefer to receive an
   amount of money for yourself and the person listed (A)
   versus an amount of money for yourself (B). Please
   circle A or B for each line.

Seven social-distance conditions consisted of the following social distances: 1, 2, 5, 10, 20, 50, and 100. Social-distance order was counterbalanced across participants. At the top of each social-distance condition page, the following instructions were shown (below is an example for the #10 person on the list):
   Imagine you made a list of the 100 people closest to you
   in the world ranging from your dearest friend or relative
   at #1 to a mere acquaintance at #100. Now imagine the
   following choices between an amount of money for you
   and the # 10 person on the list (A) and an amount for you
   (B). Circle A or B to indicate which you would choose
   in each line.

For each social distance, participants were asked to make 9 choices between a generous and selfish option. The generous option was sharing 150,000 [yen] evenly with the person at a given social-distance condition (i.e., 75,000 [yen] for the participant and 75,000 [yen] for the specific person on the list). The amount of money shared in the generous option remained unchanged across all choices. The selfish option was a hypothetical amount of money to keep exclusively for the participant, which started at 155,000 [yen] and decreased by 10,000 [yen] to 75,000 [yen] within each condition. Nine choices were on each page in two columns, with the left-hand column indicating the generous option and the right-hand column indicating the selfish option.


The last part of the questionnaire contained questions regarding participants' current frequency of USWW and their awareness of the potential danger of engaging in USWW. Prior to answering these questions, participants were told that USWW is defined as operating a smartphone while walking, with a representative example of texting a message, playing a game, watching a video, etc. First, we asked the participants whether they owned a mobile or smart phone. Second, participants were asked how many times they usually engaged in USWW per day. With regard to the frequency of USWW, Hayashi et al. (2015) asked participants to rate the frequency on a 7-point Likert scale. However, we conducted a pilot study that showed the absolute frequency of USWW that corresponds to the options on the Likert scale (e.g., "Occasionally" or "Rarely") was highly variable. Therefore, for this study we asked participants to estimate the absolute frequency value (i.e., how many times a day do you do USWW on average?). Finally, participants rated their awareness of the dangers associated with USWW on a 7-point Likert scale (i.e., how dangerous do you think USWW is?). The options of the Likert scale ranged from 1 ("not at all dangerous") to 7 ("extremely dangerous").

Data Analysis

Participant data were eliminated for the following reasons. First, 14 participants' who switched more than once between the generous and selfish options within a condition were excluded from further analyses. Second, six participants exclusively chose the delayed option during the delay discounting task, and were eliminated from analysis. Third, data from eight participants who did not answer demographic or USWW questions were eliminated. Finally, seven participants' data whose area under the curve (AUC) for delay or social discounting (as mentioned below) was shown to be an outlier by Grubbs's outlier test (Grubbs, 1969) were removed from the data set. Thus, the results reported are based on data from the remaining 204 participants: 65 females and 139 males. Data deletion criteria were based on previous delay and social discounting research (e.g., Jones & Rachlin, 2006, 2009; Romanowich & Igaki, 2017). We removed approximately 15% of the data (i.e., 35 of 239 participants) for either being nonsystematic, an outlier, or missing data. This is consistent with a recent delay discounting meta-analysis that reported average rates of non-systematic data at 18% (Smith, Lawyer, & Swift, 2018).

We selected participants for either USWW-Low or USWW-High groups according to similar criteria as Hayashi et al. (2015; see also Holt et al., 2003; Saville et al., 2010). Based on a USWW per day score [greater than or equal to] 15 (range 0-30; see Fig. 1 for details), 21 participants were classified as USWW-High. We then identified 21 participants as USWW-Low by selecting from 78 participants whose USWW frequency was either a 0,1, or 2 per day. These 21 participants selected for the USWW-Low group were chosen so that the gender distributions and mean ages were similar to the USWW-High group. Accordingly, we first selected participants for the USWW-Low group who were the same gender and age with the USWW-High group. Among those participants who remained, those who reported a lower frequency of USWW were selected for the USWW-Low group. However, this method excludes a large portion of the data (e.g., those participants in the middle of the USWW scale). Therefore, we also dichotomized USWW-Low and USWW-High groups on median USWW frequency to examine the differences between groups and retain most of the data.

Delay discounting was estimated by measuring the crossover points at which the participant switched their preference from the immediate option to the delayed option at each delay value. The crossover point was defined as the average value between the smallest amount of immediate money chosen and the amount of immediate money when the participant switched to the delayed option. For example, if a participant chose the 80,000 [yen] immediate option, but on the next choice chose the delayed option rather than the 70,000 [yen] immediate option, then the crossover point for the given delay was estimated to be 75,000 [yen] ((80,000 [yen] + 70,000 [yen])/2). For participants who exclusively chose the immediate option at a given delay, the crossover point was set to 5,000 [yen]. Thus, seven crossover points were estimated for each participant in the USWW-Low and USWW-High groups. Finally, Eq. 1 was fit to the median crossover point at each delay, creating delay discounting curves. The 100,000 [yen] and each delay value were set to A and D in Eq. 1, respectively. Nonlinear regression procedures were used to examine the goodness of fit of Eq. 1 and calculate k parameters.

Social discounting was estimated by measuring the amount of money a participant was willing to forgo in order to share with another person. Similar to delay discounting, we first calculated the crossover point at which the participant switched their preference from the selfish option to the generous option at each social distance. The crossover point was defined as the average value between the smallest selfish amount selected, and the selfish amount accepted for the first generous option. If a participant exclusively chose either the generous or selfish option at a given social distance, the crossover point was set to 160,000 [yen] and 70,000 [yen], respectively, per Jones and Rachlin (2006). Next, we calculated the amount of money the participants are willing to forgo by subtracting 75,000 [yen] from the crossover point, which was equivalent to the amount of money the participant could gain if they selected the generous option. Thus, we estimated seven amounts of money forgone for each participant in the USWW-Low and USWW-High groups. Finally, Eq. 2 was fit to the median amount of money forgone at each social distance with nonlinear regression procedures to estimate the s parameter. Because Vin Eq. 2 is defined as the undiscounted reward value at social distance = 0, the maximum amount of money forgone (i.e., 80,000 [yen] in this study) was set to V (see Romanowich & Igaki, 2017).

We also analyzed differences between USWW-Low and USWW-High groups by calculating the AUC for both delay and social discounting, which was proposed by Myerson, Green, and Warusawitharana (2001). To calculate the AUC, obtained crossover points were plotted on the vertical axis as a function of either the corresponding delay or social distance on the horizontal axis. Next, both axes were normalized by transforming the scale to a proportion of the maximum. Then, the areas of the trapezoids that were formed by the adjacent crossover points and corresponding horizontal axis were calculated. The sum of the areas of these six trapezoids represented the AUC. AUC is a value between 0 and 1, with larger AUC values indicating less discounting. Because the AUC is an atheoretical measure independent of functional form and is approximately normally distributed, it is considered a more appropriate measure for parametric statistical analyses than k and s values whose distributions are typically skewed. However, Borges, Kuang, Milhorn, and Yi (2016) showed that the conventional AUC proposed by Myerson et al. heavily overweights data at larger delays (and social distances). They proposed two alternate AUC calculations ([AUC.sub.logd] and [AUC.sub.ord]) that correct this weighting imbalance. We used the [AUC.sub.ord] calculation that transforms each delay or social distance to an ordinal scale (i.e., integers ranging from 1 through 7).

We calculated [AUC.sub.ord] for individual participants in both groups using the crossover points for both delay and social discounting. Statistical analysis was conducted for [AUC.sub.ord] using a 2 * 2 analysis of variance (ANOVA) with the type of discounting as a within-subjects factor and the degree of USWW as a between-subjects factor. Post hoc f-tests were conducted for significant main effects. Finally, in order to investigate the correlation between delay and social discounting, Pearson's correlation coefficients were calculated using individual participant [AUC.sub.ord] values for delay and social discounting.


All participants indicated that they currently owned a smartphone. Therefore, Fig. 1 shows the number of participants who self-reported different frequencies of USWW per day for all participants completing the study. USWW frequency ranged from 0 to 30 with a median number of 4 per day. There were modes at both 2 (N = 49) and 5 (N = 41) USWW per day. As mentioned above, 21 participants who reported USWW frequencies [greater than or equal to] 15 per day were selected for the USWW-High group, whereas 21 participants were selected for the USWW-Low group by reporting USWW frequencies [less than or equal to] 2.

Participant's answers on the demographic survey indicated that the sample was almost exclusively Japanese (three participants were students from outside Japan). Table 1 shows demographic information for USWW-Low and USWW-High participants. The gender distribution for both groups was not significantly different from the overall sample ([X.sub.2](1) = 0.38,p = 0.54). Furthermore, there was no significant difference between USWW-Low and USWW-High groups 0(40) = 1.26, p = 0.21) for perceived USWW danger.

Figure 2 shows the median crossover points among USWW-Low and USWW-High participants for delay (top panel) and social (bottom panel) discounting. Median crossover values decreased as an orderly function of delay magnitude and social distance. Hyperbolic functions (Eq. 1 for delay discounting and Eq. 2 for social discounting) were fit to the median crossover points for both USWW-Low (dashed line) and USWW-High (solid line) groups. The estimated exponent (k and 5) and goodness of fit for Eqs. 1 and 2 are shown in Table 2. Overall, hyperbolic functions provided a good fit to the median crossover values for both groups in terms of delay and social discounting. When visually inspecting the difference of the slopes between the USWW-Low and USWW-High groups, the slopes were steeper for the USWW-High relative to the USWW-Low participants for the delay discounting task. The slopes did not differ between groups for the social discounting task. As shown in Table 2, k and s values calculated from Eqs. 1 and 2 also support this conclusion.

Figure 3 shows the calculated [AUC.sub.ord] for both groups across delay and social discounting tasks. Before performing an ANOVA, we checked [AUC.sub.ord] normality scores using skewness and kurtosis indices. Skewness for delay discounting (USWW-Low = -0.45; USWW-High = -0.33) and social discounting (USWW-Low = 0.25; USWW-High = 0.49) were within the accepted range for a normal distribution (i.e., between -0.8 and 0.8). Furthermore, kurtosis for delay discounting (USWW-Low = -0.90; USWW-High = 0.36) and social discounting (USWW-Low = -0.02; USWW-High = -0.33) also fell into the range of normality (i.e., between -2 and 2). Therefore, we did not log transform the [AUC.sub.ord] scores. A 2 (type of discounting) * 2 (degree of USWW) ANOVA revealed a significant main effect for the type of discounting (F(1,40) = 15.27, p < 0.001, [[eta].sup.2] = 0.13) and the degree of USWW (F(1,40) = 8.03, p < 0.01, [[eta].sup.2] = 0.08). Subsequent (-tests revealed significant differences between the USWW-Low and USWW-High groups on delay discounting (7(40) = 4.06, p < 0.001, d = 1.25), but not on social discounting (7(40) = 0.38,p = 0.71, d = 0.12). The f-test results indicated that steeper discounting was observed in the USWW-High group compared with the USWW-Low group during the delay discounting task, but not during the social discounting task.

Because the criteria for the assignment of participants to either the USWW-Low or USWW-High groups discards a large portion of the data, we also divided participants into two groups on the basis of median USWW frequency (i.e., USWW = 4): USWW-Low group (0-3 times, N = 99) and USWW-High group (5-30 times, N- 101). Median value data (USWW = 4, N = 4) was excluded from the analysis. The 2 (type of discounting) * 2 (degree of USWW) ANOVA conducted for this dichotomy produced almost the same effects: a significant main effect for the type of discounting (F(1,198) = 81.73, p < .01, [[eta].sup.2] = .15), a significant main effect for the degree of USWW (F(1,198) = 7.06, p < .01, [[eta].sup.2] = .02). Subsequent 7-tests showed greater discounting from the USWW-High group compared to the USWW-Low group for the delay discounting task (7(198) = 3.19, p < .01, d = 0.45). As before, there was no difference between the groups for social discounting (7(198) = 1.05, p = .29, d = 0.15).

Figure 4 shows a scatterplot of individual participant [AUC.sub.ord] values for delay and social discounting. The Pearson's correlation coefficient indicated no systematic relationship between delay and social discounting, 7(204) = .15, p = 0.05. The dashed line through the data was fitted with linear regression ([R.sub.2] = 0.02).


The current study investigated the relationship between USWW and two types of discounting (delay and social). Hayashi et al. (2015) previously showed that TWD was positively associated with delay discounting. Therefore, we predicted that USWW would also be positively correlated with delay discounting. The present results confirmed that there is a clear positive relationship between USWW and delay discounting (see Figs. 2 and 3). In addition, we hypothesized that because USWW is dangerous for the person engaging in the behavior, as well as for people around them, USWW could be considered selfish. Therefore, we investigated whether USWW would also be correlated with social discounting. However, no significant relationship was seen between USWW and social discounting. These results suggest that USWW is closely associated with at least one aspect of impulsivity, but not with selfishness as measured by social discounting. Furthermore, the finding that USWW is correlated with delay but not social discounting is inconsistent with previous results showing that delay discounting is associated with social discounting (Rachlin & Locey, 2011). That inconsistency is also exemplified by the fact that correlational analyses of [AUC.sub.ord] within participants revealed no evidence of a relationship between delay and social discounting (see Fig. 4). The combination of no association between USWW and social discounting, and no association between delay and social discounting suggest that these two types of discounting may not be as closely related as previous research has suggested (e.g., Rachlin, 2002; Rachlin & Locey, 2011).

Similar to Hayashi et al. (2015) we found a robust association between smartphone use and delay discounting. Participants that self-reported higher frequencies of USWW discounted hypothetical rewards more steeply than participants lower USWW frequencies. This is also consistent with many previous studies examining a range of problem behaviors (e.g., Bickel & Marsch, 2001; Reynolds, 2006). The effect size between high and low frequency USWW participants was relatively large (d = 1.25), and remained moderate (d = 0.45) when conducting a median split. These effect sizes are as large, if not larger, than many clinical samples examining the association between delay discounting and drug use (MacKillop et al., 2011). This suggests a strong association that may profitably be the target of a behavioral intervention (see below for intervention suggestions).

In addition to examining the relationship between USWW and delay discounting, this study also investigated whether USWW is correlated with social discounting. When comparing the relative degree of danger between TWD and USWW, TWD may be more dangerous than USWW because car accidents can easily lead to loss of life for the person driving, as well as others around the accident. If this is the case, it can be assumed that the reason why no relationship was found between USWW and social discounting was because the degree of danger is weaker for USWW, relative to TWD. Another reason for not finding an association between USWW and social discounting may stem from what social discounting is measuring. The social-discounting questionnaire is typically limited to whether individuals keep some amount of money exclusively for themselves without sharing with others. This has been described as a measure of selfishness. However, it is unclear whether social discounting measured in this manner would pose a danger to themselves and others, which is the rationale for studying TWD and USWW. In fact, the results of this study suggest that definition may not be tenable for typical social-discounting measures. Rachlin (2002) pointed out that other than selfishness as a nonaltruistic act, there is another definition of selfishness that focuses on biological survival value. This definition led to the possibility that altruistic behavior is in itself selfish. Thus, there are multiple conceptions about the nature of selfishness. Future studies should examine how different conceptions of selfishness are correlated with each other, and predict important behaviors.

The present study provides additional data about the association between delay and social discounting. If there is an association between self-control and altruistic behavior (Rachlin & Locey, 2011), then the degree of delay and social discounting should both show the same association when compared either between groups or within participants. However, the present results indicate that there was a delay discounting difference between USWW-Low and USWW-High groups, but no group difference was observed for social discounting. Similar findings have been obtained in other studies that compared delay and social discounting in terms of cigarette smoking among pregnant women (Bradstreet et al., 2012), size of donation in a public goods game (Jones & Rachlin, 2009), and participant's BMI (Wainwright et al, 2018). We also compared individual tendencies to discount rewards for delay and social discounting with no evidence of a correlation between the two (see Fig. 4). Thus, the present study did not observe a relationship between delay and social discounting. As mentioned in the introduction, this is similar to several studies that have in general found that delay and probability discounting are independent behavioral processes (e.g., Andrade & Petry, 2012; Green, Myerson, & Ostaszewski, 1999; Holt et al, 2003; Myerson et al., 2003; Ostaszewski, Green, & Myerson, 1998).

Even though delay and probability discounting may be independent, social and probability discounting may still be related. Jones and Rachlin (2009) measured the relationship between probability and social discounting and found that participants who donated a larger amount of money in a public goods game showed shallower declines for both probability and social discounting. In addition, Rachlin and Jones (2008) showed that a reversed reward-magnitude effect also occurs for social discounting, similar to probability discounting. They suggested that there may be some link between probability and social discounting, in that the benefit a person may receive as a result of their generosity is probabilistic in nature. That is, when a person brings gifts for someone, a return gift is not always guaranteed. Although we need to be cautious about drawing any firm conclusions from these findings, additional studies should address whether probability and social discounting are correlated with each other.

The present results showed that there was no significant difference for the perceived danger of USWW between USWW-Low and USWW-High groups, even though these two groups showed a delay discounting difference. In addition, the perceived danger scores on the Likert scale were relatively high for both groups, indicating that the participants in both groups were fully aware of the danger of USWW. The tendency for participants to show distracted behavior in spite of the fact that they recognized these dangers was also found by Atchley, Atwood, and Boulton (2011) and Hayashi et al. (2015). The divergence between perceived danger and involvement in distracted behavior does not contradict our findings that the degree of USWW is correlated with delay discounting. That is, one aspect of impulsiveness is that a person can't help but persist in a harmful behavior even though they understand that doing such a behavior is harmful for themselves. If they do not realize the danger of that impulsive behavior, then that person engages in that impulsive behavior regardless of whether they are impulsive or not.

The strong relationship between USWW and delay discounting suggests that delay discounting may be an important predictor for dangerous USWW behavior. Previous studies showed that impulsivity assessed by delay discounting is useful as a predictor for substance use behavior, such as drug seeking (Carroll, Anker, Mach, Newman, & Perry, 2010) and substance-use treatment outcomes (Loree, Lundahl, & Ledgerwood, 2014). Hayashi et al. (2015) pointed out the role played by delay discounting in predicting individuals at risk for TWD. Thus, these results may make it possible to prevent accidents caused by USWW. For example, by developing a smartphone application that can measures an individual's delay discounting, the user may be sent more or less reminders as to the danger of USWW.

This study has several limitations. First, USWW was loosely defined. As noted earlier, USWW was defined as engaging in the operation of a smartphone while walking, with representative examples being texting a message, playing a game, or watching a video. Any one of these behaviors would count as a single event of USWW. However, the degree of danger differs among these behaviors. Haga et al. (2015) showed that playing games was more distracting for participants than texting messages and watching movies. Therefore, using a more concrete USWW example or parsing out individual types of USWW behavior and environment would permit a more precise evaluation of USWW and discounting. In addition, the duration for a single distracted behavior may also be important to measure. That is, even if the number of distracted behaviors are not frequent, spending a long time distracted USWW is also problematic. Moreover, the places where USWW occurs also deserve detailed consideration. When it comes to USWW in a high-traffic location such as a busy street, intersection, or station platform may be more dangerous than USWW in a low-traffic location. In fact, under the regulations with USWW in Fort Lee and Honolulu discussed in the Introduction, using smartphone "while crossing the street" is subject to a penalty. Thus, it is necessary to define USWW considering not only the frequency, but also a specific duration and location measure for USWW.

The second limitation is that USWW frequency was based on participants' self-report. These USWW self-reports may have decreased USWW accuracy. Hayashi et al. (2015) pointed out similar limitations for their TWD measure. Unlike Hayashi et al. (2015), we asked participants to estimate the absolute frequency (i.e., how many times per day they engaged in USWW), instead of using a 7-point Likert scale. Replicating Hayashi et al.'s TWD results with USWW suggests that both estimation methods are valid ways to collect self-report data. However, the absolute measure may lead to the potential for causing further problems. That is, participants have difficulty reporting precise USWW amounts, especially when the number is laige. The present findings showed that large USWW response numbers were concentrated near round numbers, such as 25 or 30 (see Fig. 1). This can be interpreted as participants' difficulty grasping an exact number of USWW when that amount is large. Unfortunately, this difficulty may have biased answers for the USWW-High group leading to an imbalance in grouping by USWW frequency. One way to address these two limitations would be to use the above-mentioned smartphone application that can more accurately detect distracted behavior. By utilizing such an application we can more accurately measure different dimensions (e.g., rate, duration, latency) of data related to smartphone operation and USWW. In addition, the location where USWW take places can also be identified using smartphones equipped with GPS function. This would help to determine whether USWW was occurring in more or less risky areas (i.e., near a busy intersection).

The third limitation is that in our primary grouping method, we selected an arbitrary number of participants for each group to maximize potential differences between USWW-High and USWW-Low groups. That is, our groups were not a representative sample from the population. Therefore, no matter how clear the difference in discounting was, it may be difficult to extrapolate the results of this study to other populations. When we used the split-median method for the secondary group comparison, the difference between USWW-High and USWW-Low for delay discounting was weaker. Future research should focus on replicating the current results using representative samples.

In sum, the present results showed differences between USWW-Low and USWW-High groups for delay discounting, but not social discounting. This suggests that one aspect of impulsivity, but not selfishness, is associated with USWW. In addition, there was no tendency for individuals who showed steeper delay discounting to show steeper social discounting. This lack of a relationship between delay and social discounting suggests that both types of discounting operate as independent processes. Future studies should examine the relationship between three types of discounting and different types of distracted behavior, in addition to smartphone use.

Compliance with Ethical Standards

All participants were informed of their rights as human participants in a psychology experiment.

Institutional Review Board acceptance at the Ryutsu Keizai University was obtained before data collection.

Conflict of Interest The authors declare no conflict of interest. No funding was used for data collection or analysis.


Alessi, S. M., & Petty, N. M. (2003). Pathological gambling severity is associated with impulsivity in a delay discounting procedure. Behavioural Processes, 64(2), 345-354.

Andrade, L. F., & Petty, N. M. (2012). Delay and probability discounting in pathological gamblers with and without a history of substance use problems. Psychopharmacology, 219(2), 491-499.

Atchley, P., Atwood, S., & Boulton, A. (2011). The choice to text and drive in younger drivers: Behavior may shape attitude. Accident Analysis & Prevention, 43(1), 134-142.

Atchley, P., & Warden, A. C. (2012). The need of young adults to text now: Using delay discounting to assess informational choice. Journal of Applied Research in Memory & Cognition, 1(4), 229-234.

Bickel, W. K., & Marsch, L. A. (2001). Toward a behavioral economic understanding of drug dependence: Delay discounting processes. Addiction, 95(1), 73-86.

Bickel, W. K., Odum, A. L., & Madden, G. J. (1999). Impulsivity and cigarette smoking: Delay discounting in current, never, and ex-smokers. Psychopharmacology, 146(4), 447-454.

Borges, A. M., Kuang, J., Milhom, H., & Yi, R. (2016). An alternative approach to calculating Area-Under-the-Curve (AUC) in delay discounting research. Journal of the Experimental Analysis of Behavior, 106(2), 145-155.

Bradstreet, M. P., Higgins, S. T., Heil, S. H., Badger, G. J., Skelly, J. M., Lynch, M. E., & Trayah, M. C. (2012). Social discounting and cigarette smoking during pregnancy. Journal of Behavioral Decision Making, 25(5), 502-511.

Caird, J. K., Johnston, K. A., Willness, C. R., Asbridge, M., & Steel, P. (2014). A meta-analysis of the effects of texting on driving. Accident Analysis & Prevention, 71, 311-318.

Carroll, M. E., Anker, J. J., Mach, J. L., Newman, J. L., & Perry, J. L. (2010). Delay discounting as a predictor of drug abuse. In G. J. Madden & W. K. Bickel (Eds.), Impulsivity: The behavioral and neurological science of discounting (pp. 243-271). Washington, DC: American Psychological Association.

Coffey, S. F., Gudleski, G. D., Saladin, M. E., & Brady, K. T. (2003). Impulsivity and rapid discounting of delayed hypothetical rewards in cocaine-dependent individuals. Experimental & Clinical Psychopharmacology, 11(1), 18-25.

Critchfield, T. S., & Kollins, S. H. (2001). Temporal discounting: Basic research and the analysis of socially important behavior. Journal of Applied Behavior Analysis, 34(1), 101-122.

Dixon, M. R., Marley, J., & Jacobs, E. A. (2003). Delay discounting by pathological gamblers. Journal of Applied Behavior Analysis, 36(4), 449-458.

Frederick, S., Loewenstein, G., & O'Donoghue, T. (2002). Time discounting and time preference: A critical review. Journal of Economic Literature, 40(2), 351-401.

Gates, S. (2013, May 17). Texting while walking ban: Fort Lee imposes $85 fines on dangerous texters [CORRECTION]. Retrieved from

Green, L., & Myerson, J. (2004). A discounting framework for choice with delayed and probabilistic rewards. Psychological Bulletin, 130(5), 769-792.

Green, L., Myerson, J., & Ostaszewski, P. (1999). Amount of reward has opposite effects on the discounting of delayed and probabilistic outcomes. Journal of Experimental Psychology: Learning, Memory, & Cognition, 25(2), 418-427.

Grubbs, F. E. (1969). Procedures for detecting outlying observations in samples. Technometrics, 7/(1), 1-21.

Hadar, A., Hadas, I., Lazarovits, A., Alyagon, U., Eliraz, D., & Zangen, A. (2017). Answering the missed call: Initial exploration of cognitive and electrophysiological changes associated with smartphone use and abuse. PLoS ONE, 12(1), e0180094.

Haga, S., Sano, A., Sekine, Y., Sato, H., Yamaguchi, S., & Masuda, K. (2015). Effects of using a smart phone on pedestrians' attention and walking. Procedia Manufacturing, 3, 2574-2580.

Hayashi, Y., & Blessington, G. P. (2018). A behavioral economic analysis of media multitasking: Delay discounting as an underlying process of texting in the classroom. Computers in Human Behavior, 86, 245-255.

Hayashi, Y., Miller, K., Foreman, A. M., & Wirth, O. (2016). A behavioral economic analysis of texting while driving: Delay discounting processes. Accident Analysis & Prevention, 97, 132-140.

Hayashi, Y., Russo, C. T., & Wirth, O. (2015). Texting while driving as impulsive choice: A behavioral economic analysis. Accident Analysis & Prevention, 83, 182-189.

Holt, D. D., Green, L., & Myerson, J. (2003). Is discounting impulsive? Evidence from temporal and probability discounting in gambling and non-gambling college students. Behavioural Processes, 64(2), 355-367.

Insurance Institute for Highway Safety (2018, March). Cellphones and texting. Retrieved from cellphonelaws

Jin, J., Pei, G., & Ma, Q. (2017). Social discounting under risk. Frontiers in Psychology, 8,292.

Jones, B., & Rachlin, H. (2006). Social discounting. Psychological Science, 17(4), 283-286.

Jones, B. A., & Rachlin, H. (2009). Delay, probability, and social discounting in a public goods game. Journal of the Experimental Analysis of Behavior, 97(1), 61-73.

Laibson, D. (1997). Golden eggs and hyperbolic discounting. Quarterly Journal of Economics, 112(2), 443-477.

Locey, M. L., Jones, B. A., & Rachlin, H. (2011). Real and hypothetical rewards in self-control and social discounting. Judgment & Decision Making, 6(6), 552-564.

Locey, M. L., & Rachlin, H. (2015). Altruism and anonymity: A behavioral analysis. Behavioural Processes, 118, 71-75.

Loewenstein, G., Read, D., & Baumeister, R. F. (Eds.). (2003). Time and decision: Economic and psychological perspectives on intertemporal choice. New York, NY: Russell Sage Foundation.

Loree, A. M., Lundahl, L. H., & Ledgerwood, D. M. (2014). Impulsivity as a predictor of treatment outcome in substance use disorders: Review and synthesis. Drug & Alcohol Review, 34(2), 119-134.

MacKillop, J., Amlung, M. T., Few, L. R., Ray, L. A., Sweet, L. H., & Munafb, M. R. (2011). Delayed reward discounting and addictive behavior: A meta-analysis. Psychopharmacology, 216(3), 305-321.

Madden, G. J., & Bickel, W. K. (Eds.). (2010). Impulsivity: The behavioral and neurological science of discounting. Washington, DC: American Psychological Association.

Madden, G. J., Bickel, W. K., & Jacobs, E. A. (1999). Discounting of delayed rewards in opioid-dependent outpatients: Exponential or hyperbolic discounting functions? Experimental & Clinical Psychopharmacology, 7(3), 284-293.

Mazur, J. E. (1987). An adjusting procedure for studying delayed reinforcement. In M. L. Commons, J. E. Mazur, J. A. Nevin, & H. Rachlin (Eds.), Quantitative analyses of behavior: Vol. 5. The effect of delay and intervening events on reinforcement value (pp. 55-73). Hillsdale, NJ: Lawrence Erlbaum Associates.

Mazur, J. E., & Biondi, D. R. (2009). Delay-amount tradeoffs in choices by pigeons and rats: Hyperbolic versus exponential discounting. Journal of the Experimental Analysis of Behavior, 91(2), 197-211.

McCartt, A. T., Kidd, D. G., & Teoh, E. R. (2014). Driver cellphone and texting bans in the United States: Evidence of effectiveness. Annals of Advances in Automotive Medicine, 58, 99-114.

McKerchar, T. L., Green, L., Myerson, J., Pickford, T. S., Hill, J. C., & Stout, S. C. (2009). A comparison of four models of delay discounting in humans. Behavioural Processes, 81(2), 256-259.

Mitchell, S. H. (1999). Measures of impulsivity in cigarette smokers and non-smokers. Psychopharmacology, 146(A), 455-464.

Mohn, T. (2017, October 23). Reading this while walking? In Honolulu, it could cost you. Retrieved from 23/business/honolulu-walking-and-texting-fine.html

Mwakalonge, J., Siuhi, S., & White, J. (2015). Distracted walking: Examining the extent to pedestrian safety problems. Journal of Traffic & Transportation Engineering, 2(5), 327-337.

Myerson, J., Green, L., Hanson, J. S., Holt, D. D., & Estle, S. J. (2003). Discounting delayed and probabilistic rewards: Processes and traits. Journal of Economic Psychology, 24(5), 619-635.

Myerson, J., Green, L., & Warusawitharana, M. (2001). Area under the curve as a measure of discounting. Journal of the Experimental Analysis of Behavior, 76(2), 235-243.

Nasar, J. L., & Troyer, D. (2013). Pedestrian injuries due to mobile phone use in public places. Accident Analysis & Prevention, 57, 91-95.

Odum, A. L. (2011). Delay discounting: I'm ak. you're a k. Journal of the Experimental Analysis of Behavior, 96(5), 427-439.

Olson, E. A., Rosso, I. M., Demers, L. A., Divatia, S., & Killgore, W. D. (2016). Sex differences in psychological factors associated with social discounting. Journal of Behavioral Decision Making 29(1), 60-66.

Ostaszewski, P., Green, L., & Myerson, J. (1998). Effects of inflation on the subjective value of delayed and probabilistic rewards. Psychonomic Bulletin & Review, 5(2), 324-333.

Parks, M. (2017, July 29). It's now illegal to text while crossing the street in Honolulu. Retrieved from 05/14/texting-while-walking-ban_n_1514308.html

Rachlin, H. (2002). Altruism and selfishness. Behavioral & Brain Sciences, 25(2), 239-250.

Rachlin, H., & Jones, B. A. (2008). Social discounting and delay discounting. Journal of Behavioral Decision Making, 27(1), 29-43.

Rachlin, H., & Locey, M. (2011). A behavioral analysis of altruism. Behavioural Processes, 57(1), 25-33.

Reed, D. D., Becirevic, A., Atchley, P., Kaplan, B. A., & Liese, B. S. (2016). Validation of a novel delay discounting of text messaging questionnaire. The Psychological Record, 66(2), 253-261.

Reynolds, B. (2006). A review of delay-discounting research with humans: Relations to drug use and gambling. Behavioural Pharmacology, 77(8), 651-667.

Romanowich, P., & Igaki, T. (2017). Effects of reward magnitude, alcohol and cigarette use on social discounting in Japan and United States college students. The Psychological Record, 67(3), 345-353.

Saville, B. K., Gisbert, A., Kopp, J., & Telesco, C. (2010). Internet addiction and delay discounting in college students. The Psychological Record, 60(2), 273-286.

Schwebel, D. C., Stavrinos, D., Byington, K. W., Davis, T., O'Neal, E. E., & de Jong, D. (2012). Distraction and pedestrian safety: How talking on the phone, texting, and listening to music impact crossing the street. Accident Analysis & Prevention, 45,266-271.

Sharp, C., Barr, G., Ross, D., Bhimani, R., Ha, C., & Vuchinich, R. (2012). Social discounting and externalizing behavior problems in boys. Journal of Behavioral Decision Making, 25(3), 239-247.

Smith, K. R., Lawyer, S. R., & Swift, J. K. (2018). A meta-analysis of nonsystematic responding in delay and probability reward discounting. Experimental & Clinical Psychopharmacology, 26(1), 94-107.

Stavrinos, D., Jones, J. L., Gamer, A. A., Griffin, R., Franklin, C. A., Ball, D.,... Fine, P. R. (2013). Impact of distracted driving on safety and traffic flow. Accident Analysis & Prevention, 61, 63-70.

Strombach, T., Jin, J., Weber, B., Kenning, P., Shen, Q., Ma, Q., & Kalenscher, T. (2014). Charity begins at home: Cultural differences in social discounting and generosity. Journal of Behavioral Decision Making, 27(3), 235-245.

Tang, Z., Zhang, H., Yan, A., & Qu, C. (2017). Time is money: The decision making of smartphone high users in gain and loss intertemporal choice. Frontiers in Psychology, 8, 363.

Tokyo Fire Department (2018). Aruki-Sumaho Nado ni Kakawaru Jiko ni Cyuui (Be careful about the accidents involved in using smartphones while walking). Retrieved from http://www.tfd.metro.

Vuchinich, R. E., & Heather, N. (Eds.). (2003). Choice, behavioural economics and addiction. Amsterdam, The Netherlands: Pergamon/Elsevier Science.

Wainwright, K., Green, B. E., & Romanowich, P. (2018). The relationship between delay and social discounting, and body mass index in university students. The Psychological Record, 65(4), 441-449.

Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

[mail] Takeharu Igaki

Takeharu Igaki (1)(iD), Paul Romanowich (2), Naoki Yamagishi (3)

(1) Faculty of Distribution and Logistics Systems, Ryutsu Keizai University, 3-2-1 Shin-Matsudo, Matsudo-shi, Chiba-ken 270-8555, Japan

(2) Department of Psychology, Gonzaga University, 502 E Boone Ave, Spokane, WA 99258, USA

(3) Department of Sociology, Ryutsu Keizai University, 3-2-1 Shin-Matsudo, Matsudo-shi, Chiba-ken 270-8555, Japan

Published online: 31 July 2019

Caption: Fig. 1. Self-reported frequency distribution of USWW per day

Caption: Fig. 2. Median crossover points for USWW-Low and USWW-High groups as a function of delays (top panel) and social distance (bottom panel). The USWW-Low group is represented by filled squares, whereas the USWW-High group is represented by open circles. The curved lines are the best-fitting discounting functions using Eq. (1) for the top panel and Eq. (2). for the bottom panel

Caption: Fig. 3. Area under the curve (AUC) for USWW-Low and USWW-High groups across delay and social discounting tasks. The USWW-Low group is represented by filled bars, whereas the USWW-High group is represented by open bars. The value for AUC is calculated by the method of Borges et al. (2016). The error bars indicate the standard error of the mean

Caption: Fig. 4. Scatterplots of Area Under the Curve (AUC) in delay discounting versus in social discounting. Each data point represents the AUC value for social discounting as a function of the AUC value for delay discounting for an individual participant The value for AUC is calculated by the method of Borges et al. (2016). The solid line was fitted with linear regression
Table 1. Demographic characteristics of the USWW-Low and
USWW-High groups

Variable                   USWW-Low       USWW-High

N                          21             21
Age in years               18.95 (0.65)   18.95 (0.65)
Gender (% male)            62%            62%
Frequency of USWW          0.81 (0.59)    20.38 (5.55)
Perceived danger of USWW   5.76 (0.87)    5.33 (1.25)

Numbers shown in parentheses are standard deviations

Table 2 Values of k from Eq. (1), s from Eq. (2) and [R.sup.2]
for USWW-Low and USWW-High groups

            Delay (Eq. 1)      Social (Eq. 2)

            k      [R.sup.2]   s      [R.sup.2]

USWW-Low    0.01   0.91        0.30   0.95
USWW-High   0.07   0.99        0.47   0.90
COPYRIGHT 2019 Springer
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Igaki, Takeharu; Romanowich, Paul; Yamagishi, Naoki
Publication:The Psychological Record
Geographic Code:9JAPA
Date:Dec 1, 2019
Previous Article:Teaching Children with Autism to Understand Metaphors.
Next Article:Symbolic Generalization of Discriminative Functions in Accordance with a Five-Member Comparative Relational Network.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters