1 Introduction

Metaphors are widely used in our daily life as a rhetorical device. Psycholinguistics suggest that metaphors are also a way of thinking to allow people to draw on concrete and familiar knowledge to reason about abstract concepts [1]. And even if metaphors are not used in sentences apparently, people could still rely on metaphors to help thinking [2]. In the domain of computer design, metaphors exhibit these properties in interfaces, they are used to produce figurative interfaces which involve a transfer of meaning from the vehicle to the topic [3]. And metaphors occur throughout the interfaces we use and design just as they invisibly permeate in our daily speech. For example, when a child learns to move a document icon from one folder to another in computer, the object and container we use in real life is actually a metaphor that will help the child to comprehend the function of folder in computer system. When working with interface metaphors, it’s essential for designers to think about which metaphors are proper and can be friendly understood by users in various operating environment.

Among all the abstract topics, time is one of the primary concepts to represent. Time is invisible and intangible, yet humans are capable of feeling time in certain ways. The temporal-spatial metaphor is a conceptual metaphor which can help construct the concept of time through spatial experience [4]. In addition, there is a broad consistency and also some differences of temporal-spatial metaphor expressions in various language and cultures [5, 6]. Therefore, it’s important to investigate the figurative interfaces of temporal-spatial metaphor in human-computer interaction design, which used in different conditions and cultures. Usually, the motion direction in space can be used to represent the motion direction of time. And there are three basic spatial axes in temporal-spatial metaphor expressions: Front-back axis, up-down axis and left-right axis. The front-back axis and up-down axis are commonly used to represent the order of time, for example, Mandarin speakers use the spatial words “qian” (“front”) and “hou” (“back”), as well as “shang” (“up”) and “xia” (“down”) to express the occurrence sequence [7]. Although left-right axis is not often used in speech to represent the temporal order, it can be found in some timing device like calendars and workflow charts [8].

When comprehending the relative motion between time and people in temporal-spatial metaphor, two kinds of time movement perspectives are often used in the front-back axis, namely the ego-moving perspective and the time-moving perspective [1, 4]. From the ego-moving perspective, time is considered static while people are in motion, and people are facing and approaching the future as in “We are coming up on Christmas”. From the time-moving perspective, people are considered static while time is in motion, and the future is approaching or upon us as in “Christmas is coming up”. Previous psycholinguistic studies have verified the psychological reality of the two perspectives of temporal-spatial metaphors [9, 10]. However, there is no systematic studies concerning the spatial schema which can represent the temporal-spatial metaphors so far. In two-dimensional space, designers always use perspective plan method to depict the distance relationship between objects. Thus, representing ego and time moving perspectives in front-back axis can rely on up-down vertical schema and left-right horizontal schema.

The present study focused on revealing the kinds of spatial schemas that could represent time movement perspectives and explored the feasibility of these spatial schemas under different situations, which will be helpful in depicting the concept of time in a human-computer interaction environment. We conducted two experiments to explore to what extent vertical and horizontal spatial schemas could represent the two perspectives of time movement. We designed a question as the reference standard to examine the validity of those schemas. The answer for the standard question comprised of a visualized picture and a descriptive sentence which could directly express the meaning of ego-moving or time-moving perspective, which was widely examined to be valid to measure perspectives of time movement in previous studies [11,12,13]. The visualized picture was referenced from the schematic of ego/time moving schema used to organize events in time in Boroditsky [11] ’s study. The descriptive sentences are matched with different time movement perspectives. Under the ego-moving perspective schema, the sentence is “I’m approaching the future”, and under the time-moving perspective schema, the sentence is “The future is approaching me”.

In experiment 1, we tested the validity of vertical spatial schema. To examine the foreshortening effect of vertical schema, we compared the foreshortening schema with a schema which contained same-size objects aligned in the up-down axis. Our assumption for experiment 1 was that the foreshortening schema would be more consistent with the standard question than the same-size vertical schema. In experiment 2, we tested the validity of horizontal spatial schema. People have no obvious preference when perceive the moving direction of elements in vertical spatial schema. However, the moving direction of elements in horizontal spatial schema was affected much more by cultural customs, such as our writing and reading direction, because most of the modern people write and read from left to right. Previous studies suggested that emotion could influence the choice of time movement perspectives. Emotion with approach motivation, such as happiness and anger, might promote the choice of the ego-moving perspective, while emotion with avoidance motivation, such as anxiety and sadness, might promote the choice of the time-moving perspective [12,13,14,15,16]. Therefore, we tried to evoke participants’ mood with one certain type of emotion of happiness, anger, anxiety or sadness. And the consistency and difference of answers between the horizontal spatial picture and the standard question would be examined. Our assumption was that in approach-motivation emotion groups (happiness and anger), the spatial picture was more consistent with the standard question than in avoidance-motivation emotion groups (anxiety and sadness).

2 Experiment 1

2.1 Participants

Participants consisted of a total of 30 undergraduate and graduate students (14 males and 16 females), ranging from 19 to 26 years of age (M = 22.7 years, SD = 1.9 years). All participants were native Chinese speakers and unaware of the purpose of the experiment.

2.2 Stimuli and Procedure

Three spatial pictures were designed as the testing stimuli (See Fig. 1): (1) Three foreshortening flying saucers aligned vertically (Fig. 1A; The flying saucer in the bottom was the largest and the one in the topside was the smallest. The picture is the ambiguous spatial target used in Boroditsky [11]’s study.); (2) Five foreshortening balls aligned vertically (Fig. 1B; The ball in the bottom was the largest and the one in the topside was the smallest.); (3) Five rectangles aligned vertically in the same size representing five soldiers in line (Fig. 1C). These three spatial pictures and the corresponding descriptions constituted the spatial schema questions.

Fig. 1.
figure 1

Three vertical spatial pictures in Experiment 1 (Note. Picture A is the ambiguous spatial target used in Boroditsky [11]’s study)

Full size image

The experiment was programmed with E-Prime 2.0 software. At the beginning of the experiment, participants were provided with an instruction for the experiment. And then they were asked to answer all the questions (Three spatial schema questions and the standard question). To avoid the order effect, the spatial schema questions and the standard question were presented randomly. For the standard question, the specific question which appeared above the schema picture was “I will attend an activity held by my school recently. Which picture can express my feeling now more properly?” The answer options were under the question. Participants were required to choose one answer from the two pictures. Answer A represents the ego-moving perspective and answer B represents the time-moving perspective.

For the vertical spatial schema A (Foreshortening flying saucers), the specific question which appeared above the picture was “There are three flying saucers arranged in a line, the biggest one is the nearest one to us, if you move one step of the middle flying saucer, which place the middle one will be in?”. If participants choose the nearest place, it means they choose the time-moving perspective, if they choose the farthest place, it means they choose the ego-moving perspective. For the vertical spatial schema B (Foreshortening balls), the specific question was “There are five balls arranged in a line, the biggest one is the nearest one to us, if you move two steps of the middle ball, which place the middle one will be in?”. If participants choose the blue ball’s place, it means they choose the time-moving perspective, and the red ball’s place represents the ego-moving perspective. For the vertical spatial schema C (Same-sized rectangles), the specific question was “There are five soldiers arranged in a line, number five is the nearest to us, if number three moves two steps, which place he will be in?” If participants choose number five, it means they choose the time-moving perspective, and number one’s place represents the ego-moving perspective.

2.3 Results

All the data were analyzed using SPSS 18.0. Firstly, for each participant, we calculated whether the answers were consistent between every vertical spatial schema question and the standard question (If both of the answers represent the same perspective, they are regarded as consistent and labelled as 1, if not, they are labelled as 0). Then we summed the number of participants for each question that labelled 1, and created a 2 × 2 contingency table between each vertical spatial schema and the standard question for ego-moving and time moving perspectives(All the contingency tables are concluded in Table 1). And we used the Kappa consistency test and the paired Chi-square test (McNemar’s test) to examine the consistency and difference of answers between the vertical spatial schema questions and the standard question.

Table 1. The participant numbers for consistent answers between vertical spatial schema and the standard question
Full size table

The results are as follows: (1) For the consistency and difference between Schema A (Foreshortening flying saucers) and the standard question, the Kappa value is 0.309, and the p value of McNemar’s test is 0.754 > 0.05; (2) For the consistency and difference between Schema B (Foreshortening balls) and the standard question, the Kappa value is 0.304, and the p value of McNemar’s test is 0.065 > 0.05; (3) For the consistency and difference between Schema C (Same-sized rectangles) and the standard question, the Kappa value is 0.082, and the p value of McNemar’s test is 0.035 < 0.05.

2.4 Discussion

Results of experiment 1 showed that the consistency between foreshortening vertical schema and the standard question was higher than the same-sized vertical schema. Accordingly, the difference between foreshortening vertical schema and the standard question was not significant, in contrast, the difference between same-sized vertical schema and the standard question was significant. Thus it indicated that the foreshortening design was crucial to characterize the time movement perspective in temporal- spatial metaphor expression.

3 Experiment 2

3.1 Participants

Participants consisted of a total of 136 undergraduate and graduate students (48 males and 88 females), ranging from 19 to 27 years of age (M = 22.6 years, SD = 1.7 years). All participants were native Chinese speakers and unaware of the purpose of the experiment. They were different students from the participants in Experiment 1. They were randomly assigned to four emotion groups (happy, angry, anxious and sad), with 34 participants in each group.

3.2 Stimuli and Procedure

The standard question mentioned above and a horizontal spatial picture (Fig. 2; five same-size matchstick men lined in the horizontal direction. The picture is referenced from Richmond [17] ’s study, the Figure b in spatial group in Appendix.) were used as the two questions to distinguish participants’ preferences of time movement perspectives. The specific question which appeared above the schema picture was “There are five matchstick men lined in the horizontal direction, if number three was moved two steps, which matchstick’s place the middle one will be in?”.

Fig. 2.
figure 2

The horizontal spatial picture in Experiment 2

Full size image

All the participants were recruited to finish the experiment in a paper-and-pencil questionnaire in a lab. After completing necessary demographic information, participants rated their current emotional states of happiness, anger, anxiousness and sadness on four nine-point Likert scales ranging from 0 (not at all) to 8 (to the extreme). They were then asked to recall and write down their own experience with one certain type of emotion of happiness, anger, anxiety or sadness. They were encouraged to write down their experience in detail within 20 min. Next, participants need to rate their feelings again on the four scales as before. Finally, all the participants were required to answer the standard question and the horizontal spatial schema question.

3.3 Results

All the data were analyzed using SPSS 18.0. Firstly, for each participant, we calculated whether the answers were consistent between the horizontal spatial schema question and the standard question in every emotion group (If both of the answers represent the same perspective, they are regarded as consistent and labelled as 1, if not, they are labelled as 0). Then we summed the number of participants that labelled 1 and created a 2 × 2 contingency table between the horizontal spatial schema and the standard question for ego-moving and time moving perspectives in each emotion group (All the contingency tables are concluded in Table 2). And we used the Kappa consistency test and the paired Chi-square test (McNemar’s test) to examine the consistency and difference of answers between the horizontal spatial schema questions and the standard question.

Table 2. The participant numbers for consistent answers between the horizontal spatial schema and the standard question
Full size table

The results are as follows: (1) In the happy group, for the consistency and difference between the horizontal spatial schema and the standard question, the Kappa value is 0.395, and the p value of McNemar’s test is 0.18 > 0.05; (2) In the angry group, for the consistency and difference between the horizontal spatial schema and the standard question, the Kappa value is 0.162, and the p value of McNemar’s test is 0.424 > 0.05; (3) In the anxious group, for the consistency and difference between the horizontal schema and the standard question, the Kappa value is −0.316, and the p value of McNemar’s test is 0.093 > 0.05; (4) In the sad group, for the consistency and the difference between the horizontal spatial schema and the standard question, the Kappa value is -0.009, and the p value of McNemar’s test is 0.003 < 0.01.

3.4 Discussion

Results of experiment 2 showed that the consistency between the horizontal schema and the standard question was different under different emotional states. In approach-motivation emotion groups, the horizontal spatial picture was consistent with the standard question in medium degree, while in avoidance-motivation emotion groups, there was no significant consistency between the horizontal spatial picture and the standard question. Previous studies implied that approach-motivation emotion would promote the choice of ego-moving perspective [12,13,14,15]. Therefore, the results indicated that the horizontal spatial schema could represent the ego-moving perspective better than the time-moving perspective.

4 General Discussion

Results from the two experiments demonstrated that the consistency between foreshortening vertical spatial schema and the standard question was much higher than the non-foreshortening schema. The horizontal spatial schema is more appropriate to describe the ego-moving perspective. The findings have implications for computer spatial schema design in time representation.

4.1 The Vertical Spatial Schema

As the findings turn out, the foreshortening design in the vertical direction is essential to express the meaning of ego-moving or time-moving perspective. Because the ego or time moving perspective reflects the relative motion between time and people in essence. It’s difficult to portray motion in the front-back axis from the human-centered angle in two-dimension space. The foreshortening design takes advantage of the perspective plan method and can help comprehending the front-back relationship in two-dimension space. And except for foreshortening objects aligned in up-down axis as the stimuli showed, foreshortening road schema might have similar effect to represent ego or time moving perspective.

4.2 The Horizontal Spatial Schema

In experiment 2, we used another idea to test the validity of the horizontal spatial schema which contains the manipulation of different emotion. No obvious foreshortening effect can be created to portray three-dimension scenes in left-right axis, so we don’t need to investigate this. The motion direction is also very important to distinguish the time movement perspective. Previous studies provided the idea that emotion plays a role in choosing time movement perspective [12,13,14,15]. Specifically, emotion with approach-motivation emotion promote the adoption of ego-moving perspective, while avoidance-motivation emotion promote the adoption of time-moving perspective. In the current study, only in approach-motivation emotion group, there is significant consistency between the horizontal spatial schema and the standard question. The approach motivation direction in left-right axis is the same as in ego-moving perspective. The findings indicated that the horizontal spatial schema is much more appropriate in ego-moving perspective representation.

4.3 The Limitation of the Study

Metaphors are ubiquity in daily speech and interface design. Metaphors serve as natural models and allow us to take our knowledge of familiar objects and events to give structure to abstract, less well understood concepts [18]. The choices of different interface metaphors can provide users with different models of the system. If the interface metaphor is not appropriate, it might cause some problems. Thus coming up with a good metaphor is essential to an easy-to-use human interface.

Our study provided two kinds of spatial schema to represent the time movement perspective, but we did not compare the effect of them directly. In future study, we will focus on the feasibility of vertical and horizontal spatial schema in real application.

In sum, the present study provided the first behavioral evidence about the representation of temporal-spatial metaphors from different spatial schemas. This result gives the valuable guidance of time representation in human-computer interaction.