Acta Psychologica Sinica

The forward testing effect in spatial route learning

MA Xiaofeng, LI Tiantian, JIA Ruihong, WEI Jie

2022, 54 (12): 1433-1442. doi: 10.3724/SP.J.1041.2022.01433

Abstract ( 403 )

HTML ( 49 )

PDF (196KB) ( 171 )

The forward testing effect describes how testing previously learned material could improve participants long-term memory for later learning of new material when continuously exposed to various information. This has been verified using different language materials. However, the effect of forward testing on spatial path learning requires further study.
This study selected 112 participants randomly and conducted two experiments to explore the forward test effect of visuospatial route learning in different directions in the same scene (Experiment 1). Further, it investigated the forward test effect of visuospatial route learning in various settings (Experiment 2). The spatial route information memory method was adopted based on the extensive experimental procedure formed by the forward test effect. Through a sequence of sites in a virtual route setting, participants were required to comprehend and recollect the structures that passed on the route. Furthermore, the exercise ended with a sequential recall test. A total of 52 participants were randomly assigned to the test and repeated study groups in Experiment 1. Eight common landmark buildings, such as hospitals and schools, were selected to form four different route information. After learning approximately 1~3 pieces of route information, the repeated study group re-learned the route information. Further, the test group recalled the order of the buildings passing through the route information as required. When learning about Route 4 regarding either the test condition or the re-learn condition, it was necessary to recall the order in which the route passed through buildings. The forward test effect of memorizing route information in different scenarios was explored in Experiment 2 with 60 participants. Unlike Experiment 1, the participants in Experiment 2 learned four different routes, each containing a different building. The experimental procedure was the same as that used in Experiment 1.
The results of experiment 1 using a 2 (group: test group, repetitive learning group) × 2 (test results: correct rate, interference rate) analysis of variance (ANOVA), which showed a significant interaction between groups and test results [F(1, 50) = 32.157, p < 0.001, η²= 0.39, see Figure 4]. Further simple effect analysis found that, the recall accuracy of spatial path information in the test group was significantly higher than in that the repeated-learning group (0.74 vs 0.32, t (50) = 5.95, p < 0.001, d = 0.64). Moreover, the active interference generated when recalling the fourth path information was considerably lower than that in the repeated-learning group (0.07 vs 0.16, t (50) = 2.831, p = 0.007, d = 0.37). The results of Experiment 2 showed that there was a positive test effect for different scene background information. 2 (groups: test group, repetitive learning group) × 2 (test results: correct rate, interference rate) analysis of variance (ANOVA) showed a significant interaction between groups and test results. The interaction between group and test results was significant [F(1, 58) = 45.483, p < 0.001, η²= 0.44, see Figure 7], the recall accuracy of spatial path information in the test group was significantly higher than in that the repeated-learning group (0.53 vs 0.24, t (58) = 5.40, p < 0.001, d = 0.57). The proactive interference in route information 4 under test condition was significantly lower than that repeated-learning condition (0.07 vs 0.27, t (58) = 5.612, p < 0.001, d = 0.59). This further proves that the application background of the forward test effect in route-information learning was extensive. More importantly, by comparing the two experimental results horizontally, it was found that different interference levels of previous information have different effects on learning following new information (Experiment 1: Figure 3 reports changes in the correct recall rate as well as the interference rate of the test group after each test in routes1-4.The recall accuracy of route 1~4 was 0.62, 0.38, 0.56, 0.74. F(3, 75) = 9.41, p < 0.001, η² = 0.27. The proactive interference rate of route 2~4 was 0.14, 0.13, 0.07. F(2, 50) = 3.28, p = 0.046, η² = 0.12. Experiment 2: The recall accuracy of route 1~4 was 0.70, 0.59, 0.73, 0.53. F (3, 87) = 4.57, p = 0.005, η² = 0.14. And the proactive interference rate of route 2~4 was 0.02, 0.04, 0.07. F(2, 58) = 4.32, p = 0.018, η² = 0.13, see Figure 6). This is manifested in the difference in the interference rate caused by the difficulty of “isolation” among materials, including the trend that the correct rate decreases when the interference rate increases and the correct rate increases when the interference rate decreases. All of these directly reveal the forward direction−the importance of counteracting proactive interference in testing the effects.
In summary, this study verified the existence of the forward test effect in the path learning of different directions in the same scene and the path learning in various settings. Extending the study of the forward testing effect on learning visuospatial path information will enrich the exploration of the forward testing effect in spatial memory. Additionally, this study found that different levels of interference from previously learned information affect the subsequent learning of new information. The findings provide direct experimental evidence for proactive interference reduction theory.

Figures and Tables | References | Related Articles | Metrics

Effects of unitization on associative and item recognition: The “benefits-only” account

LIU Zejun, GUO Chunyan

2022, 54 (12): 1443-1454. doi: 10.3724/SP.J.1041.2022.01443

Abstract ( 134 )

HTML ( 17 )

PDF (164KB) ( 54 )

It is widely accepted that unitization can promote familiarity-based associative recognition, but its effect on recognition of individual components remains unclear. A few studies have focused on this question and shown two different accounts: One is “benefits and costs” account which argues that unitization promote associative recognition at the cost of item recognition, the other is “benefits-only” account which holds that unitization can promote associative recognition without impairing item recognition. In the current study, we aimed to explore how unitization influence associative and item recognition.
Twenty-nine participants took part in the study. To avoid fatigue effects, three study-test cycles were completed with a short break (2 min) in between. For each cycle, 96 word pairs were encoded at a rate of 4 s each, with a 900~1100 ms fixation cross between trials. Forty-eight word pairs were presented in compound word pairs (CW) and 48 word pairs were presented in non-compound word pairs (NCW). After a 2-min distracting phase, participants took part in an associative recognition test, in which 64 word pairs were presented: (1) CW-intact word pairs, (2) CW-rearranged word pairs, (3) NCW-intact word pairs, and (4) NCW-rearranged word pairs. In order to matched the level of unitzation between the studied and tested word pairs, two compound word pairs were rearranged into a new compound word pairs, and the same is true of non-compound word pairs. The remaining four words were used as old stimuli in item recognition test. After all three cycles are completed, participants then took part in an item recognition test. The item test was also divided into three cycles. For each cycle, 96 single words were presented: (1) Compound-old words, (2) Non- compound-old words, and (3) new words. In both associative and item recognition tests, participants were instructed to press the “F” if the word pairs or words had been learned at encoding and to press the “J” otherwise. Meanwhile, the EEG was recorded.
First, the results showed higher level of unitization [t(28) = 44.50, p < 0.001, Cohen’s d = 8.26] and faster RTs [t(28) = −6.44, p < 0.001, Cohen’s d = −1.19] for compound word pairs than for non-compound word pairs at encoding. It indicated that the manipulation of unitization was effective in the current study. Second, concerning associative recognition, an enhanced recognition performance [t(28) = 5.67, p < 0.006, Cohen’s d = 1.05], with a larger familiarity-related FN400 effect (p = 0.015) and recollection-related LPC effect (p = 0.035), was observed for compound word pairs than for non-compound word pairs. This results suggested that unitization could improve associative recognition performance through increasing the contribution of familiarity and recollection simultaneously. And finally, an equivalent item recognition performance [t(28) = 0.97, p = 0.34] between the two word pairs was found, despite the compound word pairs elicited a larger FN400 effect than the non-compound word pairs (p = 0.049). This indicated that unitization did not impair the item recognition performance (see Fig. 1).
In summary, the current study suggests that unitization not only facilitates associative recognition but also does not impair item recognition, supporting the “benefits-only” account. Importantly, familiarity can support associative recognition when the two items were unitized into a new presentation. This means that unitization is an effective strategy for improving associative memory, especially for groups with impaired recollection.

Figures and Tables | References | Related Articles | Metrics

How state anxiety influences retrospective time duration judgment: Moderated mediating effect of cognitive appraisal and memory bias

LIU Jingyuan, LI Hong

2022, 54 (12): 1455-1466. doi: 10.3724/SP.J.1041.2022.01455

Abstract ( 217 )

HTML ( 23 )

PDF (169KB) ( 127 )

People are influenced by their emotional state and confused by environmental stimuli in anxiety, which leads to the deviation of time duration judgment. In this article, three experiments were conducted to explore the influence of state anxiety on retrospective time duration judgment, and the moderated mediating effect of cognitive appraisal and memory bias.
Experiment 1 investigated the effect of state anxiety on retrospective time duration judgment. Sixty college students participated and were randomly assigned to a high state anxiety group (n = 30, completed a procedure of anxious state induction) and a low state anxiety group (n = 30, completed a procedure of calm state induction). Then, the verbal estimation task was used to measure the retrospective time duration judgment. Experiment 2 investigated the mediating role of memory bias in the effect of state anxiety on retrospective time duration judgment through measuring memory by the free recall task. Experiment 3 investigated the moderated mediating effect of cognitive appraisal and memory bias through measuring cognitive appraisal by the visual analogue mood scales.
The results can be found below. (1) State anxiety had an effect on retrospective time duration judgment, namely, the high state anxiety individuals overestimate the duration than the low state anxiety individuals (M_high = 0.91, SD_high = 0.32; M_low = 0.72, SD_low = 0.28, t(57) = 2.43, p = 0.018, d = 0.63). (2) Memory bias played a mediating role in the relationship between state anxiety and retrospective time duration judgment (see Figure 1). (3) Cognitive appraisal moderated the mediation effect of memory bias on the influence of state anxiety on retrospective time duration judgment (coeff = 0.026, SE = 0.013, p = 0.049, 95% CI = [0.0001, 0.052], see Figure 2). Specifically, when the score of cognitive appraisal was low (M - 1 SD = 65.27), memory bias played a mediating role in the influence of state anxiety on retrospective time duration judgment (Effect = 0.042, SE = 0.032, 95% CI = [0.002, 0.131]), while when the score of cognitive appraisal was high (M + 1 SD = 93.62), memory bias did not play a mediating role in the influence of state anxiety on retrospective time duration judgment (Effect = −0.033, SE = 0.025, 95% CI = [−0.093, 0.003]).
Therefore, the effect of state anxiety on college students’ retrospective time duration judgment was a moderated mediating effect. The results reveal the internal process of the retrospective time duration judgment of anxious individuals, which can verify the attention gate model and enrich the explanatory perspective of anxiety influencing the retrospective time duration judgment through memory bias, and provide an important reference for improving the time deviation of anxious individuals through the adjustment of cognitive appraisal.

Figures and Tables | References | Related Articles | Metrics

Sleep and the consolidation of perceptual and motor sequences in implicit learning

SUN Peng, LI Xueqing, ZHANG Qingyun, SHANG Huaiqian, LING Xiaoli

2022, 54 (12): 1467-1480. doi: 10.3724/SP.J.1041.2022.01467

Abstract ( 222 )

HTML ( 17 )

PDF (359KB) ( 85 )

Implicit learning is integral to human cognition. It occurs during the learning phase (online periods) and the offline interval after the learning phase (offline periods). The process during the offline periods is referred to as consolidation, which means stabilization or enhancement of a memory trace even without additional practice after the initial acquisition. Some studies have preliminarily explored the effect of sleep on the consolidation of perceptual and motor sequences in implicit learning. However, these studies have failed to achieve a complete separation of motor sequences and perceptual sequences, thus leaving open the question of whether the sequence type moderates the effects of sleep on the consolidation of implicit sequence learning. In addition, previous studies of explicit learning have found that sequences with long length and high complexity were more likely to benefit from sleep than simple sequences, showing a sleep-based offline consolidation effect. Therefore, the question of whether the effect of sleep on offline consolidation of implicit learning of perceptual and motor sequences is moderated by sequence complexity remains unresolved.
The present study addressed these issues through three experiments applying different sequence length levels and complexities using a modified version of the Serial Reaction Time (SRT) task, which allows independent manipulation of perceptual and motor sequences. Participants were instructed to press the corresponding button as quickly and accurately as possible according to which color of the target square was the same as that of the surrounding square (Figure 1). In the perceptual sequence group, the target square color followed a sequence, but the finger response orders were randomly assigned. The opposite was true for the motor sequence group. Subsequently, a prediction test was used to estimate the amount of possible explicit knowledge.
Experiment 1 preliminarily explored the effects of sleep on the consolidation of perceptual and motor sequence implicit learning using a short six-element sequence with lower complexity. A 2 (sequence types: perceptual vs. motor) × 2 (group: day vs. night) between-subjects ANOVA showed a significant main effect of sequence types (F(1, 75) = 16.47, p < 0.001, η_p² = 0.18, Figure 2), indicating a more robust offline consolidation effect in the motor sequence group compared to the perceptual sequence group. However, sleep does not promote the offline consolidation of both sequences, F(1, 75) = 0.58, p = 0.448.
In Experiment 2, a more complex sequence (sequence length 11) was used. The results showed that participants implicitly learned the motor sequence (Figure 3). In the motor sequence group, participants with sleep performed a better offline consolidation effect than those without sleep, F(1, 71) = 12.45, p = 0.001, η_p² = 0.15. However, participants neither implicitly acquired the sequence nor showed an offline consolidation effect in the perceptual sequence group.
Given that a small or non-significant perceptual sequence learning effect were found in Experiments 1 and 2, the sleep-related offline consolidation of the perceptual sequence was further examined using a more simple sequence of length 4 in Experiment 3 (Figure 4). The results showed that participants exhibited a trend toward improvement in the performance of perceptual sequences learning, but no offline consolidation effect was observed in either group (Figure 5).
The combined results of the three experiments showed that sleep does not promote the offline consolidation of perceptual sequences, regardless of the degree of difficulty. For motor sequences, the sequence learning effect significantly increased following sleep but not after waking when the sequence length was long and structural complexity was high. However, sleep-related offline improvements were absent when the sequence length was short. In conclusion, these results indicated that the offline consolidation of implicit sequence knowledge based on sleep is modulated by sequence type and sequence complexity.

Figures and Tables | References | Related Articles | Metrics

Positive emotions enhance adaptability to contextual-cueing learning

CHEN Xiaoyu, DU Yuanyuan, LIU Qiang

2022, 54 (12): 1481-1490. doi: 10.3724/SP.J.1041.2022.01481

Abstract ( 326 )

HTML ( 25 )

PDF (317KB) ( 180 )

Contextual cueing refers to the global properties of a context or scene used to search for specific objects and regions. Chun and Jiang (1998) found that in a visual search, the reaction time to repeated configurations was shorter than the reaction time to newly generated configurations. The benefit of repeated context-target association is widely known as the contextual-cueing effect, which indicates that the subject has learned the contextual association by which attention is guided to facilitate the searching. However, the learning of contextual cueing lacks adaptability. When the subject has learned a set of contexts, it is difficult to update a new target into existing contexts (re-learning) or to learn a new set of contexts (new-learning). Previous studies have shown that restarted learning processes can facilitate the learning of new context-target associations, while updating old contexts is associated with the scope of attention. Notably, positive emotions could broaden the scope of attention and break the cognitive fixation on old processes; therefore, it is possible to improve the adaptability of contextual-cueing learning via positive emotions.
This study aimed to explore whether positive emotions could enhance the adaptability of contextual learning. To this end, we recruited a sample of 18 young adults with positive and neutral affective priming as experimental conditions and control conditions, respectively, which allowed us to explore the contextual-cueing effect under the conditions of re-learning and new-learning, the examples of re-learning and new-learning condition is shown in Figure 1, and the experimental procedure is shown in Figure 2. It should be noted that contextual cueing was defined in operation as the reaction time to the newly generated configuration minus that to the repeated configuration.
The experiment was divided into two phases: the learning phase and the switch phase (Figure 2). In Initial learning phase, subjects learned the repeated context-target associations in 3 epochs. A repeated measures ANOVA was conducted with the configuration (novel versus repeated) and the time phase (Epoch 1~3), found main effects of configuration (F(1, 17) = 46.76, p < 0.001, η² = 0.73), time phase (F(2, 34) = 22.87, p < 0.001) statistically significant, as well as the interactions between them(F(2, 34) = 4.00, p = 0.028, η² = 0.19). In post-hoc analyses, we found significant differences between configurations in every epoch. The results are shown in the Table 1 and Figure 3.
In the switch phase, the average CC and standard deviation were shown in Table 2. With the contextual-cueing effect as the dependent variable, a repeated measures ANOVA was conducted with the emotional valence (positive versus neutral), the new contextual-cueing learning type (re-learning versus new-learning), and the time phase (early phase versus late phase). It was found that the main effects of learning type (F(1, 17) = 4.57, p = 0.047, η² = 0.21) and time phase (F(1, 17) = 5.01, p = 0.039, η² = 0.23) were significant, but emotional valence (F(1, 17) = 4.31, p = 0.053) was not. The interaction among the three factors was not significant (F(1, 17) = 0.08, p = 0.783), so were emotional valence × time phase (F(1, 17) = 1.86, p = 0.191) and learning type × time phase (F(1, 17) = 4.35, p = 0.053), but the interaction between emotional valence and learning type was significant, F(1, 17) = 4.55, p = 0.048, η² = 0.21. Post-hoc analyses indicated that positive emotion only improved learning in the new-learning condition, in which the contextual-cueing effect was statistically higher in positive emotions than in neutral emotions in the late phase (Table 3 and Figure 4).
This study indicates that positive emotions can improve the adaptability of contextual-cueing learning and that the underlying mechanism is to restart the learning processing, which fails to prevent an automatic retrieval of the old presentations caused by similarity. Therefore, it facilitates the learning of new contextual cueing but does not update learned contextual cueing.

Figures and Tables | References | Related Articles | Metrics

Cognitive load and encoding methods affect prospective memory and its components in low achieving pupils in math

CHEN Youzhen, ZHANG Manman, LIN Qiurong

2022, 54 (12): 1491-1502. doi: 10.3724/SP.J.1041.2022.01491

Abstract ( 158 )

HTML ( 15 )

PDF (118KB) ( 70 )

Prospective memory is the memory for executing future intentional behavior at a proper time or occasion. Successful execution of prospective memory includes both a prospective and a retrospective component. The prospective component refers to remembering to do something when a prospective cue is encountered, and the retrospective component is the retrieval of the content of the intention to be executed. Both the prospective and retrospective components are indispensable for the successful execution of prospective memory tasks. Low achieving pupils in math with normal intelligence performed poorly on prospective memory tasks relative to high achieving math pupils. Because a failure of prospective memory may underlie academic failure in low achieving pupils in math, it is important to identify the causes of poor prospective memory performance. This study addresses the question of whether implementation intention encoding improves prospective memory performance in low achieving pupils in math and whether its effects are localized to the prospective and/or the retrospective component?
In this study, two experiments were conducted to explore the above questions. Experiment 1 used a prospective memory task that disassociated the prospective component and retrospective component. Thirty-eight (38) pupils were recruited. The study adopted a mixed design of 2 (ability group: low math achieving pupils, high math achieving pupils) × 2 (cognitive load of ongoing tasks: high, low) with the latter as a within-subjects variable. Experiment 2 investigated whether encoding conditions improved low math achieving pupils’ prospective memory. Sixty (60) low achieving pupils in math were recruited. The study adopted a mixed design of 2 (cognitive load of ongoing tasks: high, low) × 2 (encoding method: standard encoding, implementation intention encoding) with the latter as a between-subjects variable.
The results of Experiment 1 showed that accuracy rates of prospective and retrospective components of low achieving pupils in math were significantly lower than that of high achieving pupils in math, [prospective component: F (1, 34) = 5.30, p = 0.028, η_p² = 0.14; retrospective component: F (1, 34) = 21.05, p < 0.01, η_p² = 0.39]. In addition, pupils yielded significantly lower accuracy rates on the high cognitive load condition than that in the low cognitive load condition, [prospective component: F (1, 34) = 9.28, p = 0.004, η_p² = 0.21; retrospective component: F (1, 34) = 10.98, p = 0.002, η_p² = 0.25]. No significant interaction emerged between ability group and cognitive load (p > 0.05) (see Table 1).
The results of Experiment 2 replicated the above findings that significantly lower accuracy rates occurred in the high cognitive load than the low cognitive load condition, [prospective component: F (1, 54) = 7.54, p = 0.008, η_p² = 0.12; retrospective component: F (1, 54) = 11.09, p = 0.002, η_p² = 0.17]. The results also showed that the accuracy rates for the prospective and retrospective components were significantly higher for the implementation intention encoding condition than those in the standard encoding condition, [prospective component: F (1, 54) = 5.34, p = 0.025, η_p² = 0.09; retrospective component: F (1, 54) = 6.99, p = 0.011, η_p² = 0.12]. Additionally, the interaction between cognitive load and the encoding method was not significant (p > 0.05) (see Table 2).
The results indicated that low achieving pupils in math performed worse on measures of prospective memory than high achieving pupils in math. The results also showed that regardless of cognitive load, implementation intention encoding improved the performance of low math achieving pupils' prospective memory performance by enhancing both the prospective and retrospective components.

Figures and Tables | References | Related Articles | Metrics

Co-morbidity patterns of posttraumatic stress disorder and depressive symptoms: A network analysis of post-earthquake primary and secondary school students

WANG Wenchao, YUAN Hao, WU Xinchun

2022, 54 (12): 1503-1516. doi: 10.3724/SP.J.1041.2022.01503

Abstract ( 193 )

HTML ( 18 )

PDF (1765KB) ( 113 )

Post-traumatic stress disorder (PTSD) and depression have high rates of co-morbidity among primary and secondary school students who have experienced a major natural disaster. Some researchers have suggested that overlapping symptoms and dysphoria symptoms of PTSD contribute to co-morbidity, while others have attempted to explain the co-morbidity through a causal relationship between them. However, most of these studies have been based on the hypothesis of common causes, explaining co-morbidity at level of disorders or dimensions, while few studies have investigated patterns of the co-morbidity from the perspective of symptoms.
The Child PTSD Symptoms Scale (CPSS) and Center for Epidemiologic Studies Depression Scale for Children (CES-DC) were administered to two samples of primary and secondary school students one year after the earthquake (Wenchuan earthquake, N = 2530, 47.0% males, M_age = 12.86, SD = 1.96; Ya'an earthquake, N = 723, 47.7% males, M_age = 13.40, SD = 2.29). Gaussian graphical models (GGM) and Bayesian hill climbing algorithms were used to describe patterns of the co-morbidity between PTSD and depression.
It can be seen from Figure 1, overlapping symptoms and emotional numbness were the bridging symptoms. Detachment and future-limited symptoms were bridge symptoms in DSM-IV, were not bridge symptoms in the absence of DSM-IV, and fear, startle response and hypervigilance symptom were bridge symptoms. DSM-IV inaccurately defines the boundaries of PTSD, while intrusion and avoidance symptoms are core symptoms of PTSD. Figure 2 shows depressive symptoms were more likely to trigger PTSD symptoms, while intrusive symptoms triggered avoidance symptoms. The weights of each edge in the Wenchuan and Ya'an networks (all symptoms) are shown in Table 1 and Table 2, respectively.
The above findings were cross-validated in both Wenchuan and Ya'an samples, enhancing the generalizability of the findings and responding to the reproducibility crisis of psychological research. This enlightens clinical practitioners to prioritize the identification of bridging symptoms in the early assessment of clients who have suffered from traumatic events, in order to screen out clients at high-risk of co-morbid with depression. Secondly, the bridge symptoms should also be used as a breakthrough in the intervention process to develop intervention strategies. Finally, during the prognostic process, special attention should be paid to the recurrence of bridging symptoms to prevent the re-emergence of co-morbidity.

Figures and Tables | References | Related Articles | Metrics

When expectation-maximization-based theories work or do not work: An eye-tracking study of the discrepancy between everyone and every one

LIU Hong-Zhi, LI Xingshan, LI Shu, RAO Li-Lin

2022, 54 (12): 1517-1531. doi: 10.3724/SP.J.1041.2022.01517

Abstract ( 161 )

HTML ( 11 )

PDF (893KB) ( 148 )

Mainstream theorists in risky decision-making have developed various expectation theories with the ambitious goal of capturing everyone’s choices. However, ample evidence has revealed that these expectation theories could not capture every individual’s (“every one’s”) actual risky choice as descriptive theories. With doubts about the default compatibility between everyone (full set) and every one (subset), we used an eye-tracking technique to explore whether a theory for everyone would work well for every one. We found that expectation theories could capture the choice of an individual when making decisions for everyone and for self in a multiple-play condition, but could not capture the choice of an individual when making decisions for self in a single-play condition. Our findings contribute to a better understanding of the boundaries of expectation theories and those of heuristic/non-expectation models, and may shed light on the general issue of the classification of risky decision-making theories.

Figures and Tables | References | Related Articles | Metrics

How goal framing and temporal distance influence the effectiveness of COVID-19 vaccine persuasion

LIU Nan, AN Xinru, LI Aimei, LIU Pei, SUN Hailong

2022, 54 (12): 1532-1547. doi: 10.3724/SP.J.1041.2022.01532

Abstract ( 161 )

HTML ( 8 )

PDF (337KB) ( 117 )

Vaccines are crucial for controlling deadly diseases, and how to persuade people to get vaccinated has become a hot topic in enhancing public health benefits. One way to increase the vaccination rate is to raise public awareness of the importance of vaccines through advertising. As an effective and cost-friendly approach, goal framing has been widely used in vaccine advertising. However, the literature has mixed findings about whether positive or negative goal framing is more effective in persuading people to get vaccinated. The present study aims to investigate how temporal distance (present vs. future) interacts with different types of goal framing (positive vs. negative) in persuading people to get the COVID-19 vaccine. We hypothesized that negative goal framing is more persuasive when the advertising focuses on present outcomes, while positive goal framing is more effective when combined with future-focused outcomes. We further hypothesized that the inner mechanism is the intertemporal asymmetry of approach and avoidance motivation. More specifically, the avoidance motivation induced by a negative frame is stronger in the present, while the approach motivation induced by a positive frame is stronger in the future. The perceived risk of COVID-19 moderated this effect.
Four studies were conducted to examine our hypotheses. Study 1 (N = 363) was conducted to preliminarily investigate how goal framing and temporal distance jointly influence willingness to get the COVID-19 vaccine (Interaction effect: F(1, 291) = 12.25, p = 0.001, η_p² = 0.040, 90% CI [0.011, 0.083]). The results showed that a negative goal frame was more persuasive when combined with present-focused advertising, F(1, 291) = 4.42, p = 0.036, η_p² = 0.015, 90% CI [0.001, 0.046], while a positive goal frame was more effective when combined with future-focused advertising, F(1, 291) = 8.12, p = 0.005, η_p² = 0.027, 90% CI [0.005, 0.065] (Figure 1).
The aim of Study 2 (N = 292) was to verify the mediating effect of approach and avoidance motivation in a different advertising setting, as well as to rule out the potential mediators of the construal level and positive/negative emotions. The interaction effect of goal framing and temporal distance was replicated, F (1, 288) = 9.53, p = 0.002, η_p² =0.032, 90% CI [0.007, 0.072] (Figure 2). Negative goal framing was more effective in the present context, F(1, 288) = 4.22, p = 0.041, η_p² = 0.014, 90% CI[0.000, 0.045], while positive goal framing was more effective in the future context, F(1, 288) = 5.39, p = 0.021, η_p² =0.018, 90% CI [0.002, 0.052]. Avoidance motivation mediated the relationship between the goal frame and vaccine uptake in the present context, while approach motivation mediated the relationship between the goal frame and vaccine uptake in the future context (Figure 3).
In Study 3 (N = 347), we further tested the mediators by manipulating participants’ approach and avoidance motivation. The results revealed that approach motivation priming increased the persuasiveness of the present-positive frame, while future-positive frame was still more persuasive than future-negative frame, F(1, 339) = 11.12, p = 0.001, η_p² =0.032, 90% CI [0.008, 0.068]. On the other hand, avoidance motivation priming increased the persuasiveness of the future-negative frame, while present-negative frame was still more persuasive than present-positive frame, F(1, 339) = 20.93, p < 0.001, η_p² = 0.058, 90% CI [0.024, 0.103], see Figure 4.
Study 4 (N = 423) was a quasi-experiment in which we recruited participants from areas with different levels of COVID-19 risk to test how perceived risk moderated the interaction effect of goal framing and temporal distance. The results indicated that when the COVID-19 risk was low, present-negative frame was more effective than present-positive frame, F(1, 415) = 6.45, p = 0.011, η_p² = 0.015, 90% CI [0.002, 0.040]; and future-positive frame was more effective than future-negative frame, F(1, 415) = 7.62, p = 0.006, η_p² = 0.018, 90%[0.003, 0.044], see Figure 5a. The same pattern as in the former studies. However, when the COVID-19 risk was high, the difference in vaccine uptake between present-positive and present-negative conditions disappeared, F(1, 415) = 0.31, p = 0.579, while the future-positive frame was still more persuasive than the future-negative frame, F(1, 415) = 3.93, p = 0.048, η_p² = 0.009, 90% CI [0.000, 0.031] (Figure 5b).
In conclusion, the present study found an interactive effect of goal framing and temporal distance in persuading people to get the COVID-19 vaccine. Avoidance/approach motivation mediates the relationship between goal framing and vaccine uptake in the present/future temporal context. The perceived COVID risk further moderated the interaction effect. The present study contributes to both the framing and approach-avoidance motivation literature and sheds light on future practices in persuading people to get the COVID vaccine and promoting the uptake of other vaccines.

Figures and Tables | References | Related Articles | Metrics

Subjective social class positively predicts altruistic punishment

CHEN Sijing, YANG Shasha, WANG Hao, WAN Fenghua

2022, 54 (12): 1548-1561. doi: 10.3724/SP.J.1041.2022.01548

Abstract ( 173 )

HTML ( 16 )

PDF (144KB) ( 142 )

Altruistic punishment means that people privately bear the cost to punish norm violators, although the punishment yields no material gain. The positive effects of altruistic punishment on cooperation and norm maintenance are well documented and the possible mechanisms underlying these effects have also been widely tested. However, an important issue remains underexplored: Does people’s social background influence their altruistic punitive behavior? If yes, how? Across four studies, this article tested the relationship between altruistic punishment and social class, potential psychological mechanisms underlying this relationship, as well as some boundary conditions.
Study 1 used the Chinese general social survey (2013) released by the National Survey Research Center at Renmin University of China to examine the relationship between altruistic punishment and social class. We selected two items as the dependent variables of Study 1 (D13: employees reported environmental pollution at their own cost; D23: employees retaliated against their foreign boss who insulted China). After screening the samples, a total of 4921 (for D13) and 4864 (for D23) valid data were obtained, respectively. The results showed that after controlling for educational attainment and annual income, participants’ subjective social class significantly positively predicts their altruistic punishment (D13: B = 0.07, Wald = 16.70, OR = 1.08, 95% CI [1.04, 1.11], p < 0.001; D23: B = 0.05, Wald = 8.74, OR = 1.06, 95% CI [1.02, 1.09], p = 0.003).
Study 2 was a real-life event-based survey with 450 participants. In Study 2, we distinguished direct altruistic punishment from indirect altruistic punishment and further investigated the differential effects of social class on them. The results revealed a positive effect of subjective social class on direct altruistic punishment (Wald = 8.50, OR = 1.32, 95% CI [1.10, 1.59], p = 0.004), but not on indirect altruistic punishment (Wald = 1.14, OR = 1.10, 95% CI [0.92, 1.33], p = 0.286).
Study 3 was a 2 (social class: low/high) × 2 (punishment cost: low/high) between-participants design, and the main purpose was to demonstrate the moderating role of punishment cost in the process of social class affecting altruistic punishment. A 2 × 2 between-participants ANOVA indicated a significant main effect of social class (F(1, 228) = 6.96, p = 0.009, η_p² = 0.03), with upper class being more likely to punish violators overall, and a significant main effect of cost (F(1, 228) = 20.09, p < 0.001, η_p² = 0.08), with participants overall having higher levels of punishment in the low-cost condition; and more importantly, the interaction between the two was also significant (F(1, 228) = 4.90, p = 0.028, η_p² = 0.02). The results of simple effects analysis were shown in Figure 1. In the low-cost condition, no significant difference was found in punishment between high-class (M = 24.93, SE = 1.33) and low-class participants (M = 24.37, SE = 1.30) (F(1, 228) = 0.09, p = 0.763); while in the high-cost condition, compared to high-class participants (M = 21.93, SE = 1.33), low-class participants (M = 15.52, SE = 1.32) punished significantly less (F(1, 228) = 11.67, p < 0.001).
Based on the survey data, Study 4 (N = 125) proposed a conditional process model with belief in a just world (BJW) as mediator and punishment cost as moderator, hereby providing an explanatory framework for the impact of social class on altruistic punishment. The results of the cross-level mediation model showed that BJW positively predicted altruistic punishment (B = 2.42, SE = 0.76, p = 0.001) and that the predictive effect of the social class remained significant after the inclusion of BJW (B = 1.05, SE = 0.47, p = 0.026). The test of the cross-level moderating effect of the punishment cost revealed a significantly negative interaction term between BJW and the punishment cost (B = −0.90, SE = 0.27, p = 0.001), as well as a significantly positive interaction term between social class and the punishment cost (B = 0.44, SE = 0.14, p = 0.002). Table 1 depicts the direct effect of social class on altruistic punishment, the indirect effect through BWJ, and the total effect, at three cost levels (M ± 1 SD). Taken together, these results demonstrated that social class affects altruistic punishment indirectly mainly through belief in a just world when punishment cost is low, whereas social class directly affects altruistic punishment when punishment cost is high.
To sum up, we have found evidence that upper-class (vs. lower-class) individuals are more willing to engage in altruistic punishment in economic games and real-life contexts, implying that in a modern society increasingly stratified along class lines, people’s social background should not be ignored in the research of altruistic punishment. In addition, the results of this article also prove that on the one hand, altruistic punishment is at least partly a non-strategic sanction, because an important driver that triggers altruistic punishment is to protect their just belief, and on the other hand cost-benefit based considerations are not completely absent in altruistic punishment.

Figures and Tables | References | Related Articles | Metrics

Failed players, successful advertisements: Does showing the failure experience increase observers’ intention to try?

LUAN Mo, LI Junpeng

2022, 54 (12): 1562-1578. doi: 10.3724/SP.J.1041.2022.01562

Abstract ( 286 )

HTML ( 26 )

PDF (245KB) ( 146 )

It is a common strategy of advertising to show pleasant experience of users, but the effect may be opposite on game advertising. Most games are based on the principle of competition, which makes the result of the game a key factor. Success or failure would not only affect the motivation of game players, but also affect the observers. However, there are few researches on investigating how observing others' failure influences individuals' willingness to try a task, especially in the context of game. Would observing the failure (vs. success) of others in the ads of game improve observers' intention to download the game? If yes, what are the underlying psychological mechanism and boundary conditions? Based on social comparison theory and competition theory, the current research explored the influence of observing others' failure on observers' download intention of the games, and the serial mediation model of downward social comparison and competitive motivation, as well as the moderating role of difficulty and the observers' trait competitiveness were discussed.
The results of Experiment 1 provided evidence for the influence of game results on download intention. Experiment 1A (N = 310) was a 3 (game results: success vs. failure vs. control) between-subjects design, which proved that participants observing the failure (M_failure= 74.19, SD = 26.35) were more likely to download the game than those observing the success (M_success= 62.27, SD = 19.00; p = 0.001, η_p² = 0.06) and those in the control condition (M_control = 56.44, SD = 27.64; p < 0.001, η_p² = 0.12). Experiment 1B tried to replicate the main effect in a real-world context. Two visions of a game ad (success vs. failure) were put in a short video social platform. Before Experiment 1B, we conducted a pretest to ensure that our manipulation of game result was effective and did not differ in terms of credibility and understandability. Results of the analysis revealed that a consumer who viewed the failure ad (2.12%) was more likely to click on the ad than a consumer who viewed the success ad (1.24%; χ²(1, N = 9993) = 11.48, p < 0.001).
Experiment 2 (N = 250) measured downward social comparison and competitive motivation with a 2 (game result: success vs. failure) between-subjects design through another game ad. We hypothesized that the results of the game influence observers’ download intention through the serial mediator effect of downward social comparison and competitive motivation. A pretest was conducted to ensure our manipulation of game result was effective and did not differ in terms of credibility, understandability and difficulty. Results of Experiment 2 proved the serial mediator effect of downward social comparison and trait competitiveness in the impact of observing others’ failure on download intention (indirect effect = 0.33, 95% CI: [0.16, 0.58]; Figure 1), which verified hypothesis 2.
Experiment 3 (N = 250) was 2 (game result: success vs. failure) ×2 (difficulty: simple vs. difficult) between-subjects design. In experiment 3, we devised two levels in a same game that varied in their difficulty. Results of Experiment 3 suggested that for simple tasks, observing failure facilitated observers’ download intention (M_failure= 6.98, SD = 1.74; M_success= 5.19, SD = 2.53; p < 0.001, η_p² = 0.15; Figure 2), whereas for difficult tasks, this effect was no longer significant (M_failure= 6.21, SD = 2.39; M_success= 6.11, SD = 2.35; p = 0.808). In other words, task difficulty played a moderating role between observing failure and download intention, which confirmed hypothesis 3 and further verified the mediating effect.
Experiment 4 (N = 250) explored the moderating role of observers’ trait competitiveness with a 2 (game result: success vs. failure) between-subjects design through the same ad of Experiment 1A. Results of Experiment 4 supported that the observer's trait competitiveness played a moderating role between observing failure and observers’ download intention (F (1, 246) = 6.07, p = 0.014). For participants with high trait competitiveness (+1 SD), observing the failure of others would lead to higher download intention (p < 0.001). For participants with low trait competitiveness (−1 SD), this effect was no longer present (p = 0.241). The results of Johnson-Neyman analysis (Figure 3) showed that observing failure increased download intention for participants who scored > 4.37 on the 7-point trait competitiveness scale (Effect = 7.87, p = 0.05). Hypothesis 4 was supported.
Taken together, based on the game advertising situation, these studies confirmed the positive effect of observing others' failure on observers’ behavior intention, and expanded the research on the impact of displaying failed product experience on advertising in the field of consumer behavior. The findings of the current research also added an alternative perspective to the social learning literature.

Figures and Tables | References | Related Articles | Metrics

Archive