The Experiment

What Are We Building?

We are building an online resurgence experiment: a classic three-phase resurgence preparation, delivered as a task that participants complete in their own web browser. It is a human analogue preparation for studying the conditions under which previously suppressed behavior returns.

From the participant's side, it is one short, self-contained session. They consent, answer a few demographic questions, and read the task instructions. Then the task itself: colored circles move around the screen, and clicking them sometimes earns points. Which circle earns, and when it stops earning, changes across three phases, and participants are never told the rules or when they change. Whether responding returns to the original circle in the final phase is the measurement. The session ends with a debrief explaining what the study was about.

How that session gets delivered to real participants (where the experiment runs, where the data goes, how people are recruited and paid) is the next lesson's topic. This lesson is about the science: what resurgence is, why the design has three phases, and which variables the preparation lets you manipulate.

What Is Resurgence?

Resurgence is defined as the increase in a previously suppressed behavior following a relative worsening of conditions for a more recently reinforced alternative behavior (Lattal et al., 2017; Shahan & Craig, 2017; Nist & Shahan, 2021). Resurgence has been studied extensively across nonhuman animals, typically developing humans, and clinical populations.

Resurgence is a reliable and replicable phenomenon, demonstrated across many settings, response topographies, and subject populations, and is repeatable within individual subjects (Kestner et al., 2018; Cook & Lattal, 2019). This robustness makes it well-suited for online preparations.

The Three-Phase Experimental Setup

The experiment follows the standard three-phase resurgence design:

Phase 1: Acquisition (Target Response Reinforced)

A target response (R1) is reinforced, establishing a behavioral history for that response. The duration and schedule of reinforcement in this phase are known to influence subsequent resurgence.

Phase 2: Differential Reinforcement of Alternative Behavior (DRA)

The target response (R1) is placed on extinction, and an alternative response (R2) is introduced and reinforced. This phase serves to suppress R1. Key variables in this phase that influence resurgence include:

  • Rate of alternative reinforcement: Higher rates of alternative reinforcement during Phase 2 produce greater suppression of R1 and are associated with greater resurgence in Phase 3 (Shahan & Sweeney, 2011; Helvey et al., 2023; Podlesnik & Kelley, 2014)
  • Phase 2 duration: Evidence is mixed; some studies show longer Phase 2 reduces resurgence (Leitenberg et al., 1975), while others find no significant effect (Winterbauer et al., 2013; Nall et al., 2018), as reviewed by Smith and Greer (2022)
  • Degree of suppression: Incomplete suppression of R1 during Phase 2 is often associated with failure to observe resurgence in Phase 3 (Smith & Greer, 2022)

Phase 3: Resurgence Test

Alternative reinforcement is reduced or eliminated, and the dependent variable is whether R1 returns. This phase operationalizes resurgence. The increase in target responding is transient but reliable (Kestner et al., 2018). Critically, resurgence should occur whenever alternative reinforcement is worsened relative to Phase 2, regardless of the specific source or schedule of that alternative reinforcement (Shahan & Sweeney, 2011).

Why These Design Decisions?

Why Prolific?

Prolific provides access to a diverse, pre-screened participant pool with high data quality relative to other crowdsourcing platforms. Human analogue resurgence preparations have been successfully run online, and the platform allows for precise control over inclusion criteria (e.g., device type, age, first language) that are relevant to task performance.

Why a Computer-Based Button-Press Task?

Human analogue resurgence tasks using button-press or point-earning paradigms have been validated across multiple laboratories and shown to produce results consistent with nonhuman preparations (Kestner et al., 2018; Ritchey et al., 2023). These tasks are well-suited to online deployment because they require no specialized equipment and can be administered and completed in a single session.

Why Three Phases?

The three-phase design is the standard experimental structure for studying resurgence and maps directly onto clinically relevant sequences: skill acquisition, behavioral intervention (DRA), and treatment lapse or reinforcement thinning (Shahan & Sweeney, 2011; Craig et al., 2018). This structure also provides the clearest test of resurgence as distinct from other relapse phenomena such as renewal or reinstatement.

Why Study Resurgence in Humans Online?

Understanding the variables that modulate resurgence has direct implications for the design of durable, relapse-resistant behavior interventions, particularly DRA-based treatments used in applied behavior analysis (Greer & Shahan, 2019; Shahan et al., 2020). Online human preparations allow for rapid, large-scale data collection that can complement and extend findings from laboratory and clinical settings. Resurgence has been shown to be both replicable across subjects and repeatable within subjects, making it well-suited to within-subject designs that are feasible in online research (Kestner et al., 2018).

Variables of Interest

Depending on the specific research question, the following variables may be manipulated or measured across conditions:

VariableRole
Rate of alternative reinforcement (Phase 2)Independent variable
Phase 2 durationIndependent variable
Reinforcer magnitudeIndependent variable
Response effortIndependent variable
Stimulus contextIndependent variable
Target response rate (Phase 3)Primary dependent variable

References

  • Bachá-Méndez, G., Reid, A. K., & Mendoza-Soylovna, A. (2007). Resurgence of integrated behavioral units. Journal of the Experimental Analysis of Behavior, 87(1), 5–24. https://doi.org/10.1901/jeab.2007.55-05

  • Cook, J. E., & Lattal, K. A. (2019). Repeated, within-session resurgence. Journal of the Experimental Analysis of Behavior, 111(1), 28–47. https://doi.org/10.1002/jeab.496

  • Craig, A. R., Browning, K. O., & Shahan, T. A. (2017). Stimuli previously associated with reinforcement mitigate resurgence. Journal of the Experimental Analysis of Behavior, 108(2), 139–150. https://doi.org/10.1002/jeab.278

  • Craig, A. R., Cunningham, P. J., Sweeney, M. M., Shahan, T. A., & Nevin, J. A. (2018). Delivering alternative reinforcement in a distinct context reduces its counter-therapeutic effects on relapse. Journal of the Experimental Analysis of Behavior, 109(3), 492–505. https://doi.org/10.1002/jeab.431

  • Craig, A. R., & Shahan, T. A. (2016). Behavioral momentum theory fails to account for the effects of reinforcement rate on resurgence. Journal of the Experimental Analysis of Behavior, 105(3), 375–392. https://doi.org/10.1002/jeab.207

  • Fujimaki, S., Hu, T., & Kosaki, Y. (2024). Resurgence of goal-directed actions and habits. Journal of the Experimental Analysis of Behavior, 121(1), 97–107. https://doi.org/10.1002/jeab.884

  • Helvey, C. I., Fisher, W. W., Greer, B. D., Fuhrman, A. M., & Mitteer, D. R. (2023). Resurgence of destructive behavior following differential rates of alternative reinforcement. Journal of Applied Behavior Analysis, 56(4), 804–815. https://doi.org/10.1002/jaba.1010

  • Kestner, K. M., Diaz-Salvat, C. C., St. Peter, C. C., & Peterson, S. M. (2018). Assessing the repeatability of resurgence in humans: Implications for the use of within-subject designs. Journal of the Experimental Analysis of Behavior, 110(3), 545–552. https://doi.org/10.1002/jeab.477

  • Kimball, R. T., Kelley, M. E., Podlesnik, C. A., Forton, A., & Hinkle, B. (2018). Resurgence with and without an alternative response. Journal of Applied Behavior Analysis, 51(4), 854–865. https://doi.org/10.1002/jaba.466

  • Neely, L., Graber, J., Kunnavatana, S., & Cantrell, K. (2020). Impact of language on behavior treatment outcomes. Journal of Applied Behavior Analysis, 53(2), 796–810. https://doi.org/10.1002/jaba.626

  • Nist, A. N., & Shahan, T. A. (2021). Resurgence and repeated within-session progressive-interval thinning of alternative reinforcement. Journal of the Experimental Analysis of Behavior, 115(2), 442–459. https://doi.org/10.1002/jeab.672

  • Podlesnik, C. A., & Kelley, M. E. (2014). Resurgence: Response competition, stimulus control, and reinforcer control. Journal of the Experimental Analysis of Behavior, 102(2), 231–240. https://doi.org/10.1002/jeab.102

  • Ritchey, C. M., Kuroda, T., & Podlesnik, C. A. (2023). A quantitative analysis of resurgence following downshifts in alternative-reinforcer magnitude. Journal of the Experimental Analysis of Behavior, 119(3), 501–512. https://doi.org/10.1002/jeab.843

  • Shahan, T. A., & Sweeney, M. M. (2011). A model of resurgence based on behavioral momentum theory. Journal of the Experimental Analysis of Behavior, 95(1), 91–108. https://doi.org/10.1901/jeab.2011.95-91

  • Smith, S. W., & Greer, B. D. (2022). Phase duration and resurgence. Journal of the Experimental Analysis of Behavior, 117(1), 91–104. https://doi.org/10.1002/jeab.725

  • Walter, K. M., & Dickson, C. A. (2023). Response effort and resurgence. Journal of the Experimental Analysis of Behavior, 119(2), 373–391. https://doi.org/10.1002/jeab.835