Todd Kashdan in his Psychology Today blog recently wrote about the nuanced relationship between gratitude and trust and I encourage readers to read that post as well as the study by Kashdan and colleagues on which the blog post is based.
I read the study and the blog post with much interest as it adds substantially to our knowledge base of how gratitude leads to trust and how the relationship is partially mediated by positive affect.
Basically, Kashdan and colleagues used an online variant of reflecting on 'three good things daily' gratitude intervention (in their variant, one reflects on five good things over the past three days, and for three moments in time over the span of a week) to boost gratitude among the experimental group and they operationlized trust using the money returned in the Trust Game (a variant of the Dictator game) that was conducted in laboratory.
Some of their basic findings can be summarized as follows:
These are all pretty good findings that add substantially to our knowledge, however the interpretation of some of these by Kashdan seems rather strange to me.
To start with, he assumes that no difference at baseline means that there was no difference in social interactivity readiness or looking-forwardness between the two groups and this he equates with no difference in trust levels at baseline. However, by reading the paper its not apparent whether the participants knew it would be a social experiment, all one can make out is that they were told they will participate in a financial experiment. This may appear pedantic, but there is a long research tradition showing that economic and moral domains are sometimes orthogonal and if people were primed to think about money that may have reduced both their social propensities as well as positive affect. I would rather tell the participants they will be participating in a social experiment during baseline, if I wanted to draw any conclusions about baseline condition.
Another point I would have liked to be addressed whether the gratitude intervention did indeed increase state gratitude (is there a state gratitude measure that could have been administered?). One indirect way to measure it would have been to administer a state measure of affect say PANAS before the laboratory task to see if gratitude intervention had indeed led to more positive affect and thus more gratitude feelings. Anyway these are minor squabbles.
The major disagreement I have is with regards to interpretation of the physiological findings; the paper itself mentions that the physiological finding could be due to enthusiasm rather than stress; the blip that starts once the trust game stars and continues during the game could be due to enthusiasm and excitement at interacting with a stranger, rather than stress at making a decision say under uncertainty; those with more grateful feelings were excited at an opportunity to help someone or trust someone once the trust game was told to them and then onward showed more excitement, enthusiasm and positive affect; that at least is my interpretation of the findings and this does not merit a conclusion that gratitude or trust may have a complicated relationship with health.
Lastly, I would have liked to see more emphasis on the finding that experimental group that trusted more also felt more grateful and perhaps got caught in a a upward virtuous circle where gratitude leads to positive affect leads to trust leads to altruistic actions leads to positive affect leads to gratitude and so on....
But of course simplistic models like this go against the grain of anti-parsimoniousness!
Overall I think its a very important piece of work, and its quite presumptuous of me to have a different interpretation than one of the study authors, but by challenging interpretations and takeaways, that's how I believe science progresses.
Drążkowski, D., Kaczmarek, L.D., & Kashdan, T.B. (in press). Gratitude pays: A weekly gratitude intervention influences monetary decisions, physiological responses, and emotional experiences during a trust-related social interaction. Personality and Individual Differences