ʻŌhiʻa (Metrosideros polymorpha) is an endemic keystone tree species critical to the structure and function of Hawaiian forests. Rapid ʻŌhiʻa Death (ROD), caused by fungal pathogens, has killed millions of ʻōhiʻa trees on Hawaiʻi Island. The unprecedented rate of tree mortality has fueled a multi-agency collaboration in response to ROD. Preventing the spread of this disease requires both resident and tourist engagement in various behaviors; for example, inhibiting movement of infected soils. In 2016, Hawaii DLNR began installing boot brush stations at Na Ala Hele Hawaiʻi Island trailheads. Signs posted on the station explain ROD and encourage trail users to engage in behaviors that reduce the potential spread of fungal spores that transmit ROD.
Despite the widespread use of signs in parks and on trails to encourage good user behavior, research shows that signage is often ignored or at worst, can promote opposite behaviors (Cialdini et al., 2013). A large body of social science research indicates the power of social norms in influencing behavior. Injunctive norms indicate what behavior people "ought" to do, versus descriptive norms which state what behaviors individuals usually engage in. Both have been demostrated to be effective in influencing behavior. In studies of signage affecting trail user behavior, however, proscriptive (negatively framed) injunctive normative messages led to the highest rate of desired behavior engagement (Cialdini et al., 2006, 2013; Winter et al. 2006).
In this exploratory pilot project, we looked at the influence of various signage on trail user behavior. We examined the effectiveness of five different messages aimed at promoting the use of a boot brush station at Pololū Valley (4 tests +control). To mobilize action in the context of ROD prevention, we focused on various forms of normative social influence. Each of the 4 tested signs represented a frame discussed in prominent social science research (descriptive norms (general & localized), injunctive norm, fear/threat). The messages on the signs variously appealed to different motivating aspects of human behavior, including fear, environmental attitudes, or social norms.
Photo: Boot brush station at Pololū Valley trailhead
Our team utilized participant observation to see if hikers were using the boot brush station placed at Pololū Valley trailhead. We observed a total of 1,037 individuals who hiked the trail. We tracked engagement from each sign through 30 total hours of observation, which took place on five separate Fridays from 8 AM - 2 PM. One day was designated for each of the five signs, which carried our experimentally derived messages. We selected Pololū Valley trail because it is a high traffic trail that is frequented by both residents and tourists. We kept tallies of individuals that used the boot brush station, those that read the sign but did not use the boot brush station, individuals that did not see the boot brush station at all, and those that glanced at the boot brush but did not engage in the desired behavior.
Table 1: Five types of signs were tested in this experiment
Sign Type | Message |
Sign 1: Standard environmental message (control) | Stop the spread of Rapid ʻŌhiʻa Death! Help Protect Hawaii’s native forests! |
Sign 2: Injunctive norm (proscriptive) | Please do not proceed down this trail without using this boot brush! |
Sign 3: Descriptive social norm- Pololu Valley | Stop the spread of Rapid ʻŌhiʻa Death! Join 61% of Pololū Valley trail users that have started using a boot brush station. |
Sign 4: Threat message | Are you a carrier? In the past two weeks, have you [done the desired behaviors]? If so, then you might be carrying ROD. |
Sign 5: Descriptive social norm- Hawaii Island | Stop the spread of Rapid ʻŌhiʻa Death! Join 61% of Hawaiʻi Island residents that have started using a boot brush station. |
Sign 1: Standard environmental message (control)
Sign 2: Injunctive norm (proscriptive)
Sign 3: Provincial social norm (localized to trail)
Sign 4: Threat/fear message
Sign 5: Descriptive social norm (generalized to island)
The control sign employed a standard environmental message (Help Protect Our Forests) with the desired behaviors. It was the least effective sign with only 5.1% of hikers choosing to use the boot brush station. The regression analysis also revealed that the control sign was significantly and negatively associated with boot brush usage when compared to the other messages. The second sign was red and offered a clear proscriptive (negative frame -"do not") injunctive message, “Please do not proceed down this trail without using this boot brush station.” It was the most effective sign, with 46.3% of hikers using the booth brush. When we performed a linear regression and left out the injunctive norm sign, all other signs were negatively associated with boot brush usage. The third sign was a social norms sign specific to Pololū Valley, which highlighted the percentage of people using a boot brush station (Percentage was determined through a short social media survey asking about engagement in ROD-prevention behaviors). About a quarter (24.4%) of trail users used the boot brush station with this sign. The fourth sign was red and employed a threat message (“Are you a carrier...you might be carrying ROD”). When the threat sign was posted, approximately 18% of hikers used the boot brush station. The fifth and final sign was, like sign #3, a social norms sign, however it was generally focused on Hawaii Island trails instead of a specific trailhead. This sign led to an 18% rate of trail users engaging in the desired behavior.
Table 3: Coefficients and standard errors linking different sign types with engagement in boot brush use using a binary logistic regression
Predictor | Used boot brush |
Test #1 Control: Standard environmental message Red/Please: Injunctive norm Pololū Valley: Provincial social norm Red/Boot print: Threat/fear message |
-0.63 (0.41) 2.14 (0.31)** 1.16 (0.34)** 0.78 (0.33) * |
Test #2 Red/Please: Injunctive norm Pololū Valley: Provincial social norm Red/Boot print: Threat/fear message HI Island: Descriptive social norm |
2.78 (0.32)** 1.80 (0.35)** 1.41 (0.35)** 0.63 (0.41)** |
Test #3 Pololū Valley: Provincial social norm Red/Boot print: Threat/fear message HI Island: Descriptive social norm Control: Standard environmental message |
-0.99 (0.22)** -1.37 (0.21)** -2.14 (0.31)** -2.78 (0.32)** |
Test #4 Red/Boot print: Threat/fear message HI Island: Descriptive social norm Control: Standard environmental message Red/Please: Injunctive norm |
-0.38 (0.26) -1.16 (0.34)** -1.79 (0.35)** 0.99 (0.22)** |
Test #5 HI Island: Descriptive social norm Control: Standard environmental message Red/Please: Injunctive norm Pololū Valley: Provincial social norm |
-0.78 (0.33)* -1.41 (0.35)** 1.37 (0.21)** 0.38 (0.26) |
*p < .05
**p < .01
While all of the various messages increased engagement when compared to the control sign, the results of the injunctive-proscriptive sign were striking. Nearly half (46.3%) of trail users utilized the boot brush when this sign was posted, compared to a fraction (5%) with the control. When adjusted to account only for users who were observed to actually notice the boot brush station at all, the usage was actually more than 75%, indicating that better placement of the station would likely increase engagement even more.
While the proscriptive-injunctive message led the most engagement, as predicted by the literature, the descriptive norms signs and threat message signs also increased engagement. Similar to findings by Goldstein et al. 2008, the more proximate the message to the physical setting (this specific trail vs. all island trails), the higher the rate of behavior. The generalized social norm sign and the threat message sign were equal in engagement rate.
Increased engagement may also be creating a positive feedback loop through social modeling. Anecdotally, our observers reported that whenever trail users were actively using the station to clean their boots when new hikers arrived, the likelihood that the newcomers would use the boot brush increased.
Our results supported our hypothesis that when a proscriptive-injunctive message was presented to the hikers, they were more likely to engage in using the boot brush than when a standard environmental message, descriptive norm messages, or threat messages were displayed. Future expansion of this project includes observations at different high traffic trailheads in Hawaiʻi and the inclusion of a proposed prescriptive-injunctive message.
Cialdini, R. B. (2013), “Crafting Normative Messages to Protect the Environment,” Current Directions in Psychological Science, 12(4), 105-109.
Cialdini, R. B., Demaine, L. J., Sagarin, B. J., Barrett, D. W., Rhoads, K., & Winter, P. L. (2006), “Managing social norms for persuasive impact,” Social Influence, 1(1), 3-15.
Goldstein, N. J., R. B. Cialdini, and V. Griskevicius (2008), “A room with a viewpoint: Using social norms to motivate environmental conservation in hotels,” Journal of Consumer Research 35 (3):472–482.
Winter, P.L. 2006. The impact of normative message types on off-trail hiking. Journal of Interpretation Research, 11 (1): 35-52.