Limits...
The Measurement of Subjective Value and Its Relation to Contingent Valuation and Environmental Public Goods.

Khaw MW, Grab DA, Livermore MA, Vossler CA, Glimcher PW - PLoS ONE (2015)

Bottom Line: The activations associated with the traditional classes of goods replicate previous correlations between neural activity in valuation areas and behavioral preferences.In contrast, CV-elicited values for environmental proposals did not correlate with brain activity at either the individual or population level.For a sub-population of participants, CV-elicited values were correlated with activity within the dorsomedial prefrontal cortex, a region associated with cognitive control and shifting decision strategies.

View Article: PubMed Central - PubMed

Affiliation: Center for Neural Science, New York University, New York City, New York, United States of America.

ABSTRACT
Environmental public goods--including national parks, clean air/water, and ecosystem services--provide substantial benefits on a global scale. These goods have unique characteristics in that they are typically "nonmarket" goods, with values from both use and passive use that accrue to a large number of individuals both in current and future generations. In this study, we test the hypothesis that neural signals in areas correlated with subjective valuations for essentially all other previously studied categories of goods (ventromedial prefrontal cortex and ventral striatum) also correlate with environmental valuations. We use contingent valuation (CV) as our behavioral tool for measuring valuations of environmental public goods. CV is a standard stated preference approach that presents survey respondents with information on an issue and asks questions that help policymakers determine how much citizens are willing to pay for a public good or policy. We scanned human subjects while they viewed environmental proposals, along with three other classes of goods. The presentation of all four classes of goods yielded robust and similar patterns of temporally synchronized brain activation within attentional networks. The activations associated with the traditional classes of goods replicate previous correlations between neural activity in valuation areas and behavioral preferences. In contrast, CV-elicited values for environmental proposals did not correlate with brain activity at either the individual or population level. For a sub-population of participants, CV-elicited values were correlated with activity within the dorsomedial prefrontal cortex, a region associated with cognitive control and shifting decision strategies. The results show that neural activity associated with the subjective valuation of environmental proposals differs profoundly from the neural activity associated with previously examined goods and preference measures.

No MeSH data available.


Related in: MedlinePlus

(a) Regions correlated with environmental preferences (p < 0.05, uncorrected) for exploratory sub-population (orange) relative to a priori vmPFC region of interest (green). (b) Selection of exploratory sub-population based on average whole-brain regression coefficient for environmental proposals. (c) Differential activity within exploratory sub-population’s vmPFC region for favored vs. nonfavored proposals. (d) Reaction time differences between exploratory sub-population and remaining subjects.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4519262&req=5

pone.0132842.g009: (a) Regions correlated with environmental preferences (p < 0.05, uncorrected) for exploratory sub-population (orange) relative to a priori vmPFC region of interest (green). (b) Selection of exploratory sub-population based on average whole-brain regression coefficient for environmental proposals. (c) Differential activity within exploratory sub-population’s vmPFC region for favored vs. nonfavored proposals. (d) Reaction time differences between exploratory sub-population and remaining subjects.

Mentions: To investigate more deeply the relationship between behavioral valuations elicited by the CV procedure and neural activity, we performed a within-subject regression on CV-elicited valuations and selected all subjects who showed a positive (though not necessarily significant) correlation globally (averaged across all voxels) for each whole brain (Fig 9b). Once these individuals were identified (N = 12), the same random-effects regression that was performed previously on the whole sample was re-run within this sub-population to test for statistically significant correlations. Using this admittedly unconventional procedure, we were able to identify a sub-population of subjects that exhibited a positive correlation (p < 0.05, uncorrected; though one must be cautious in interpreting this statistical significance given the post-hoc selection process) in a single region of the brain (Fig 9a). Surprisingly, this correlation between environmental valuations and brain activity was observed not in a valuation-associated area but rather in the dorsomedial prefrontal cortex (dmPFC). Indeed, using the same percent change analysis employed earlier, the vmPFC signal for this sub-population did not yield significantly differential activity between most and least-preferred environmental goods (Fig 9c). This finding suggests that there are, as expected, brain activations associated with the CV-elicited preferences, but it strengthens the conclusion that these correlations do not lie within traditional valuation areas. It should also be noted that at a post-hoc behavioral level, this group exhibits longer CV-trial completion times ( = 6.83, s.d. = 0.09) compared to the remainder of our sample ( = 6.23, s.d. = 0.06, Fig 9d). Subsequent to the completion of this analysis, a median split of the subjects using the response time (RT) distribution (selecting for the slower half of subjects, with average completion times greater than Md = 6.55 s) also yielded a similarly significant correlation between environmental valuations and activity in the dmPFC (p < 0.05, uncorrected) for this sub-population. It is important to note, however, that this supplementary analysis was performed as a subsequent validation exercise and thus should not be interpreted as a second piece of independent evidence for a sub-population effect.


The Measurement of Subjective Value and Its Relation to Contingent Valuation and Environmental Public Goods.

Khaw MW, Grab DA, Livermore MA, Vossler CA, Glimcher PW - PLoS ONE (2015)

(a) Regions correlated with environmental preferences (p < 0.05, uncorrected) for exploratory sub-population (orange) relative to a priori vmPFC region of interest (green). (b) Selection of exploratory sub-population based on average whole-brain regression coefficient for environmental proposals. (c) Differential activity within exploratory sub-population’s vmPFC region for favored vs. nonfavored proposals. (d) Reaction time differences between exploratory sub-population and remaining subjects.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4519262&req=5

pone.0132842.g009: (a) Regions correlated with environmental preferences (p < 0.05, uncorrected) for exploratory sub-population (orange) relative to a priori vmPFC region of interest (green). (b) Selection of exploratory sub-population based on average whole-brain regression coefficient for environmental proposals. (c) Differential activity within exploratory sub-population’s vmPFC region for favored vs. nonfavored proposals. (d) Reaction time differences between exploratory sub-population and remaining subjects.
Mentions: To investigate more deeply the relationship between behavioral valuations elicited by the CV procedure and neural activity, we performed a within-subject regression on CV-elicited valuations and selected all subjects who showed a positive (though not necessarily significant) correlation globally (averaged across all voxels) for each whole brain (Fig 9b). Once these individuals were identified (N = 12), the same random-effects regression that was performed previously on the whole sample was re-run within this sub-population to test for statistically significant correlations. Using this admittedly unconventional procedure, we were able to identify a sub-population of subjects that exhibited a positive correlation (p < 0.05, uncorrected; though one must be cautious in interpreting this statistical significance given the post-hoc selection process) in a single region of the brain (Fig 9a). Surprisingly, this correlation between environmental valuations and brain activity was observed not in a valuation-associated area but rather in the dorsomedial prefrontal cortex (dmPFC). Indeed, using the same percent change analysis employed earlier, the vmPFC signal for this sub-population did not yield significantly differential activity between most and least-preferred environmental goods (Fig 9c). This finding suggests that there are, as expected, brain activations associated with the CV-elicited preferences, but it strengthens the conclusion that these correlations do not lie within traditional valuation areas. It should also be noted that at a post-hoc behavioral level, this group exhibits longer CV-trial completion times ( = 6.83, s.d. = 0.09) compared to the remainder of our sample ( = 6.23, s.d. = 0.06, Fig 9d). Subsequent to the completion of this analysis, a median split of the subjects using the response time (RT) distribution (selecting for the slower half of subjects, with average completion times greater than Md = 6.55 s) also yielded a similarly significant correlation between environmental valuations and activity in the dmPFC (p < 0.05, uncorrected) for this sub-population. It is important to note, however, that this supplementary analysis was performed as a subsequent validation exercise and thus should not be interpreted as a second piece of independent evidence for a sub-population effect.

Bottom Line: The activations associated with the traditional classes of goods replicate previous correlations between neural activity in valuation areas and behavioral preferences.In contrast, CV-elicited values for environmental proposals did not correlate with brain activity at either the individual or population level.For a sub-population of participants, CV-elicited values were correlated with activity within the dorsomedial prefrontal cortex, a region associated with cognitive control and shifting decision strategies.

View Article: PubMed Central - PubMed

Affiliation: Center for Neural Science, New York University, New York City, New York, United States of America.

ABSTRACT
Environmental public goods--including national parks, clean air/water, and ecosystem services--provide substantial benefits on a global scale. These goods have unique characteristics in that they are typically "nonmarket" goods, with values from both use and passive use that accrue to a large number of individuals both in current and future generations. In this study, we test the hypothesis that neural signals in areas correlated with subjective valuations for essentially all other previously studied categories of goods (ventromedial prefrontal cortex and ventral striatum) also correlate with environmental valuations. We use contingent valuation (CV) as our behavioral tool for measuring valuations of environmental public goods. CV is a standard stated preference approach that presents survey respondents with information on an issue and asks questions that help policymakers determine how much citizens are willing to pay for a public good or policy. We scanned human subjects while they viewed environmental proposals, along with three other classes of goods. The presentation of all four classes of goods yielded robust and similar patterns of temporally synchronized brain activation within attentional networks. The activations associated with the traditional classes of goods replicate previous correlations between neural activity in valuation areas and behavioral preferences. In contrast, CV-elicited values for environmental proposals did not correlate with brain activity at either the individual or population level. For a sub-population of participants, CV-elicited values were correlated with activity within the dorsomedial prefrontal cortex, a region associated with cognitive control and shifting decision strategies. The results show that neural activity associated with the subjective valuation of environmental proposals differs profoundly from the neural activity associated with previously examined goods and preference measures.

No MeSH data available.


Related in: MedlinePlus