IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 13, NO. 1, MARCH 2012 71 Subliminal Persuasion and Its Potential for Driver Behavior Adaptation Andreas Riener Abstract—Mental overload is a problem drivers are increasingly exposed to in today’s complex task of vehicle operation and is one of the causes of traffic accidents or hazards. To keep road safety high but allow for additional information to be forwarded to the driver, we propose to employ subliminal persuasion: a technique where the information is transferred below the level of conscious awareness. Thus, the driver becomes aware of the information, but his/her cognitive load is unaltered. To analyze the potential of this approach, we have designed a case study implementing an “eco- driving” strategy operating in the background. Driving economy is thereby estimated based on vehicles’ mileage gathered in real time from numerous sensors in and around the car, and information is conveyed to the driver with very light, not attentively perceivable, vibration patterns originating from tactor elements integrated into the safety belt or the car seat. The main research hypothesis followed in this paper and investigated in real driving studies is that drivers would operate their vehicles more economically on vibrotactile instructions perceived inattentively, as compared with the case without any notifications. Indeed, results indicate an improvement in driving economy for segments driven with sublim- inal feedback compared with routes driven without assistance but not without qualifications. Statistical significance has been proven for the safety belt interface, whereas it has not been substantiated for the tactile car seat. (However, more research is needed to validate the applicability of subliminal persuasion across a wider range of driving and in-vehicle tasks.) Index Terms—Driving economy, safety belt interface, sublimi- nal persuasion, tactile driver seat, vibrotactile stimulation. I. POTENTIALS IN ECO-FRIENDLY DRIVING T O DRIVE economically has recently become one of the most discussed topics in the car domain, and not only because of the commercial crisis and the demands of envi- ronmental authorities. Picking up on this current topic, “eco- driving,” in the sense used here, addresses two aspects, namely, fuel consumption usage and CO 2 emissions. We start with a re- view of state-of-the-art and recent developments for both facets and highlight future saving potentials. Starting in 1999, the car manufacturer Volkswagen produced the “Lupo 3L” in series, consuming as little as 3 L of fuel per 100 km. Further advance- ments in engine technology, power train, chassis materials, etc., Manuscript received February 28, 2011; revised June 29, 2011, September 26, 2011, and October 3, 2011; accepted October 5, 2011. Date of publication January 7, 2012; date of current version March 5, 2012. This work was supported by the Future and Emerging Technologies Programme within the Seventh Framework Programme of the European Commission under Grant 231288 (SOCIONICAL). The Associate Editor for this paper was M. Á. Sotelo Vázquez. The author is with the Institute for Pervasive Computing, Johannes Kepler University, 4040 Linz, Austria (e-mail: riener@pervasive.jku.at). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TITS.2011.2178838 have recently made the dream of 2-L cars a reality [1]. However, for unknown reasons, running cars still consume on average 6–7 L of fuel per 100 km [2]—twice to triple the amount required from a technological point of view. With increasing fuel prices, the trend for fuel-saving vehicles, which translates into lower running costs and money saved by drivers, has be- come evident. This is impressively demonstrated at automotive exhibitions such as the Frankfurt trade fair (IAA), where the releases of midget and small cars with low fuel consumption (“green cars”) have attracted much attention, whereas informa- tion booths for heavy fuel-consuming vehicles, if on display at all, were left empty (observed in 2009). Lower running cost is one factor accounting for the ready sale of green cars, but another reason for their production is the societal and governmental demand to reduce CO 2 emissions. The European Commission, for example, has enacted a regulation for new cars sold in the EU-27 to reach the 120-g CO 2 /km target on average emission by 2012 (IP/07/155, February 7, 2007), which is a reduction of around 25% from 2007 levels. The U.S. government is on the way to adopt similar restrictions based on fuel economy [2]. Electric or hybrid vehicles are different approaches to adhere to a more economic style of driving. Unfortunately, electric cars, due to their drawback of high initial costs, limited performance, and excessively short range, can only be used in niche applications. Hybrid vehicles, on the other side, partially overcome the limitations of electric cars (short range) and fuel-driven vehicles (high dependence of engine operation efficiency on traffic conditions) by using an electric motor in situations with bad engine efficiency [3] and changing to a common gas/diesel engine for long traveling distances. Although reductions can reach, in urban driving with frequent acceleration/deceleration phases and many short stops, 20%–30% [4], the average savings are projected to be much lower and are highly dependent on vehicle usage or routes driven (hybrid vehicles when driven interurban degen- erate to common motor vehicles with zero savings). For both approaches to be of value with respect to CO 2 emission savings, electricity not produced by burning fossil fuels is another issue to consider [4]. Due to the drawbacks, electric/hybrid vehicles will no longer be considered here. Individual Options: Not only advances in technology but also the driver itself can contribute to economic driving. In addition to the general option to reach reductions by changing the type of vehicle used or reducing the distances driven, the style of driving also carries potential. Reference [5] showed that accelerating more moderately is a feasible approach to reduce CO 2 emissions. The U.S. Federal government has stated that “the way you drive can affect fuel economy by up to 33%” [6], and Liimatainen [7] reported that driving behavior is one of the 1524-9050/$31.00 © 2012 IEEE