Delay discounting of different outcomes: Review and theory
Food, sex, and drugs lose value faster than money, and how you set up the choice matters as much as the reward itself.
01Research in Context
What this study did
Odum and his team read every delay-discounting paper they could find. They compared how people and animals choose money, food, sex, or drugs when the reward is delayed.
They looked at 40 years of data to see which rewards lose value fastest and why.
What they found
Food, sex, and drugs lose value faster than money. A slice of pizza today beats two slices next week, but five dollars today beats five dollars next week less often.
Discounting is partly a trait (stable like eye color) and partly a state (shifts with hunger, stress, or mood).
How this fits with other research
Nickerson et al. (2015) showed that making people sit and wait at the computer steepens discounting for cigarettes. Odum’s review says the same thing: procedure details change the curve.
McKerchar et al. (2019) found that double-reward questions fit the math, but variable-delay questions do not. Odum folds this in: how you ask matters as much as what you ask about.
Holt et al. (2018) proved pigeon key-pecks give clean discounting data. Odum uses this animal evidence to show the pattern holds across species.
Davis et al. (1972) and Corrigan et al. (1998) both showed longer unsignaled delays crush response rates. Odum links these old parametric studies to the new discounting curve work.
Why it matters
If a client chooses a tiny immediate reinforcer over a bigger delayed one, check the reward type. Swap money tokens for edible treats and the curve may steepen. Also, keep delays short and signal them clearly—old basic science says unsignaled waits kill responding. You can reshape discounting for one reward, and the change may spread to others, so target the client’s favorite first.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Signal every delay and start with money-like tokens before shifting to edible rewards.
02At a glance
03Original abstract
Steep delay discounting is characterized by a preference for small immediate outcomes relative to larger delayed outcomes and is predictive of drug abuse, risky sexual behaviors, and other maladaptive behaviors. Nancy M. Petry was a pioneer in delay discounting research who demonstrated that people discount delayed monetary gains less steeply than they discount substances with abuse liability. Subsequent research found steep discounting for not only drugs, but other nonmonetary outcomes such as food, sex, and health. In this systematic review, we evaluate the hypotheses proposed to explain differences in discounting as a function of the type of outcome and explore the trait- and state-like nature of delay discounting. We found overwhelming evidence for the state-like quality of delay discounting: Consistent with Petry and others’ work, nonmonetary outcomes are discounted more steeply than monetary outcomes. We propose two hypotheses that together may account for this effect: Decreasing Future Preference and Decreasing Future Worth. We also found clear evidence that delay discounting has trait-like qualities: People who steeply discount monetary outcomes steeply discount nonmonetary outcomes as well. The implication is that changing delay discounting for one outcome could change discounting for other outcomes.
Journal of the Experimental Analysis of Behavior, 2020 · doi:10.1002/jeab.589