1 Introduction

Rapidly growing energy use is exhausting energy sources and resulting in energy cost increases and has heavy environmental impacts, including ozone layer depletion, global warming, and the climate change. In developed countries, the building sector accounts for more energy consumption (38 % of nation’s total energy use [1] ) than the transportation sector and the industry sector which includes manufacturing, agriculture, and mining. Within the building sector, office buildings present opportunities in energy reduction due to their large share in energy consumption (18 %) [2] as well as the fact that occupants in commercial buildings are not in charge of energy bills and therefore, they are not motivated to take actions in reducing energy costs. Considering the increasing trends in buildings’ energy consumption, efficient energy usage in buildings plays a vital role in reducing global energy demand and associated emissions. Therefore, energy efficiency in buildings is a prime objective for energy policy at the regional, national, and international levels [2]. Existing approaches to increase energy efficiency in buildings fall into two broad categories: technological improvements and advancements in building systems (e.g., HVAC systems, lighting systems, sensors and sensing systems) and behavioral techniques (e.g. modifying behavior of an occupant by providing feedback (e.g., personal energy use)). Energy efficient occupant behavior provides higher savings with much lower costs than the investments in building equipment and it can be used in both new and existing buildings [3]. Previous work showed that occupant actions and behaviors have a significant impact (more than 40 %) on building energy demand [4]. In order to reduce office buildings energy consumption, this research focuses on novel human-building communication methods to influence occupants’ behavior positively towards more energy efficient choices.

Our review of relevant research in the energy domain revealed some missing elements, which result in the failure of existing behavior change interventions to fully induce the possible changes in human behavior. In addition, follow-up studies conducted in energy domain to monitor the effects of the interventions over longer periods of time showed that the positive effects of the intervention were not maintained after the interventions were stopped for a while [5]. We argue that human-building communication needs continuous interaction. In fact, our main goal is to transform buildings into interactive living spaces that communicate with their occupants via agents and influence the way the occupants interact with their building to enable energy efficiency. As a first step towards this goal, we investigated effective communication methods aimed at influencing building occupants’ energy-related behaviors.

2 Hypothesis

Social influence strategies in other domains (health, marketing, psychology) introduced approaches that have been successful in enhancing the effectiveness of intervention strategies that influence behavior. Existing influence strategies in energy domain failed to fully utilize social features. Social features are related to how humans interact with each other, requiring another person or an agent (in our study). Our objective is to investigate the effects of incorporating social influence methods (behavior-change tactics more commonly seen in face-to-face communication such as foot in the door, rule of reciprocity, and direct request) into the occupant-building communication. We hypothesized that social influence methods adopted in the design of our persuasive messages will influence the users’ compliance with the energy saving messages.

3 Methodology

As the first step toward enhancing the human-building communication, we examined the use of social influence methods, such as direct requests and compliance-gaining techniques. Compliance gaining refers to the interactions, in which one individual (the agent) attempts to induce another person (the target) to perform a desired behavior that the target person otherwise might not have performed [6]. We tested the compliance-gaining strategies that have the greatest influence on behavior in the marketing literature, which are the (a) foot-in-the-door, and (b) role of reciprocity techniques [79]. In addition, we tested the effects of directly asking the participants to engage in the desired behavior, which is called (c) direct request.

To test the influence of these social influence methods in the context of office buildings – occupant communication, Immersive Virtual Environments (IVEs) are used to simulate a real-life office setting. Although performing such experiments is possible in existing buildings, there are several factors that could affect the results (e.g., cloudy/sunny weather in different days). IVEs give us the ability to manipulate complex, abstract objects and concepts while maintaining high experimental control. They also allow us to control for potentially confounding variables that exist in built environments and isolate the variables of interest. We designed an office environment in 3ds Max© and imported it to Unity 3D© to be used in an IVE.

The participants first were assigned randomly to one of the three groups of social influence strategies: a) foot-in-the-door, (b) role of reciprocity techniques, and (c) direct request (Table 1).

Table 1. Messages delivered to participants in different groups

The messages were delivered through text and the participants were given an opportunity to comply with the message in the IVE. First, the participants were instructed to adjust the lighting levels (intensity levels) of the room to their most preferred settings by opening the blinds or turning the artificial lights on/off. Through real-time rendering, the virtual model dynamically adjusted the lighting levels as the participants turn the light switch on/off or opened/closed the blinds. Once the participants acknowledged their most preferred setting and were trained how to interact with the environment, they were immersed in the main scene where all the blinds were closed and the artificial lights were on. Then participants were exposed to different persuasive messages according to the condition to which they were assigned while performing a given task and then were given an opportunity to comply with different requests. The participants were asked to comply with the request by adjusting the lighting setting in an office and perform a set of activities, such as watching a video in the modeled environment using a head-mounted-display.

We observed the participants’ compliance and, in a post survey, we ask the participants to explain the reasons behind their actions and to rate their intentions to use a similar suggestion system, if it were employed in their office in the future. Our intention was to understand how similar systems could be employed in the design of future buildings and operation of existing buildings. Additionally, the participants were asked to fill out a pre survey (personality test) and post surveys (group ideology survey and Technology Readiness Index) to conduct exploratory analyses on the effect of the participant attitudes and personalities with the way that they might interact with the suggestion system (Fig. 1).

Fig. 1.
figure 1

Messages delivered in the IVE

4 Results

The results are presented based on the three social influenced methods commonly used in social science. Data were collected during an ongoing experiment that included 32 participants (59 % male and 41 % female). The participants were recruited from the graduate students in the University of Southern California. In general, 70 % of the participants complied with the message and the results indicated that among the social influence methods that were tested, role of reciprocity received the highest rate of compliance (Fig. 2).

Fig. 2.
figure 2

Participants’ compliance percentages in different groups

When participants were asked how likely they were to use a similar suggestion system, if it were employed in their office, the results showed that 69 % of the participants rated the possibility as vey likely and somewhat likely, 10 % as undecided, and 21 % as unlikely or somewhat unlikely.

5 Conclusion and Discussion

In this study, we investigated the influence of incorporating social influence methods to improve information-based energy focused persuasive messages. It is an essential first step towards enhancing communication by making it more social. Our approach differs from previous work in the energy domain, in which the information-based persuasive messages were used or energy consumption feedback was provided through charts/tables (usually without another agent). The results of the test showed that the social influence methods that apply principal of reciprocity (rule of reciprocity) received more compliance followed by the methods that adapt repetitive requests (foot in the door) in comparison with the direct requests which neither use the principle of reciprocity nor repeated requests.