Complete documentation for researchers and administrators using the Paced Auditory Serial Addition Task — Computerized (PASAT-c).
The Paced Auditory Serial Addition Task — Computerized (PASAT-c) is a computerized behavioral measure of distress tolerance: the capacity to persist at a goal-directed task while experiencing emotional distress and frustration. It was developed by Lejuez, Kahler, and Brown (2003), who adapted the original Paced Auditory Serial Addition Task (PASAT; Gronwall, 1977) — a neuropsychological test of attention and information processing — into a controlled laboratory stressor.
Where the original PASAT was designed to assess cognitive functioning, the PASAT-c deliberately exploits the task's well-documented capacity to provoke stress, irritability, and frustration, turning that property into a precise instrument for studying how long a person will tolerate distress before choosing to quit.
Participants are shown single digits one at a time and must add each new digit to the digit that immediately preceded it, then click the sum on an on-screen keypad. A running score is displayed and updated continuously, and incorrect or missed responses trigger an unpleasant auditory tone that adds to the frustration. Difficulty escalates across three levels by progressively shortening the interval between digits. The transition from Level 1 to Level 2 happens seamlessly, without warning; a brief rest separates Levels 2 and 3, during which self-report ratings can be collected. Most importantly, during Level 3 participants are given an explicit Quit option and may end the task whenever they wish.
| Level | Interval between digits | Difficulty | Approx. duration |
|---|---|---|---|
| Level 1 | 3.0 seconds | Low | 3 minutes |
| Level 2 | 1.5 seconds | Medium | 5 minutes |
| Level 3 | 1.0 second | High | Up to 10 minutes (Quit option available) |
The PASAT-c's primary outcome is Level 3 persistence — the latency to quit, i.e., how long a participant continues under the hardest condition before choosing to stop. This behavioral index of distress tolerance is the task's signature contribution. The platform also captures several secondary measures:
The PASAT-c is used widely as a behavioral marker of distress (in)tolerance, a transdiagnostic vulnerability implicated in many forms of psychopathology. Representative research areas include:
The original PASAT (Gronwall, 1977) was created to assess information processing and sustained attention after head injury, but clinicians repeatedly observed that it provoked marked stress, anxiety, and irritability — a feature long considered a nuisance for neuropsychological assessment. Lejuez and colleagues (2003) recognized that this same property made the task valuable as a standardized stressor, and added computerized stimulus timing, an explicit escape option, and built-in self-report and auditory-feedback features to create the PASAT-c. The construct it measures — distress tolerance — has since become the focus of a large clinical literature (e.g., Zvolensky, Bernstein, & Vujanovic, 2011).
This guide walks you through setting up a research study using the PASAT-c platform, from creating an account to distributing the task to participants.
To begin using the platform, you'll need a researcher account:
From your dashboard, you can create a customized PASAT-c experiment:
The platform is being built to support two deployment options for different research needs:
Browser version (web-based):
Desktop application (planned):
Once your experiment is configured, you can distribute it using several methods:
Method 1 — Direct code distribution. Share the generated experiment code (e.g., PSY-2401) with participants, who enter it on the home page under "Participant Access." Ideal for in-person laboratory studies.
Method 2 — Direct link. Generate a direct URL from your dashboard and distribute it by email, survey platform, or participant portal. The link loads the correct experiment configuration automatically.
Method 3 — QR code. Generate a QR code to display in the lab or on recruitment materials; participants scan it to open the task.
This section provides the theoretical background, standard parameters, scoring, and interpretation considerations for the PASAT-c, drawn from the primary literature. It is intended to help researchers administer and analyze the task in a way that is consistent with published work.
Distress tolerance refers to the capacity to withstand aversive internal states and to persist at goal-directed behavior while distressed (Zvolensky, Bernstein, & Vujanovic, 2011). It is treated as a transdiagnostic individual difference: low distress tolerance is associated with smoking, substance use, disordered eating, and relapse, and its reduction is an explicit target of several cognitive-behavioral treatments (McHugh et al., 2011). The PASAT-c operationalizes the behavioral, persistence-based facet of this construct. Rather than asking people how well they think they tolerate distress, it places them under escalating, frustrating demand and records whether — and when — they choose to escape.
The configuration most frequently used in the original laboratory is summarized below. Each parameter can be adjusted to suit a study's needs, but deviations should be reported.
| Phase | Interval | Duration | Key feature |
|---|---|---|---|
| Level 1 | 3.0 s | 3 min | Low difficulty |
| Level 2 | 1.5 s | 5 min | Seamless, unannounced transition from Level 1; medium difficulty |
| Rest break | — | ~2 min | Self-report ratings collected |
| Warning | — | 15 s | "Get ready" / "get set" / "go" |
| Level 3 | 1.0 s | Up to 10 min | High difficulty; explicit Quit option |
To encourage genuine effort without creating a ceiling on persistence, the original protocol offered a modest incentive (a small reward for scoring above the group average). Stimuli are kept arithmetically trivial — no sum exceeds 20 — so that mathematical ability does not confound the measure.
The headline measure is latency to terminate Level 3 — the elapsed time before a participant clicks Quit (Lejuez et al., 2003). Longer latencies indicate greater distress tolerance. A characteristic feature of PASAT-c data is that a substantial subset of participants never quit and complete the full Level 3; in the original pilot sample, roughly one-third did so. Because of this, latency-to-quit distributions are typically non-normal, and the original authors recommended analyzing them with survival analysis (treating non-quitters as censored observations), an appropriate statistical transformation, or a dichotomous quit/no-quit grouping.
One methodological subtlety deserves attention. In the original data, better task performance was associated with a lower risk of quitting, raising the possibility that proficiency (attention and motor speed) partly drives persistence. However, self-reported distress predicted quitting independently of performance — for example, higher irritability was associated with greater risk of termination even after controlling for performance — which supports the view that persistence and performance are at least partly separable. Researchers should therefore record performance alongside persistence and take it into account when interpreting results.
The PASAT-c converges with other behavioral distress-tolerance tasks, such as the computerized Mirror-Tracing Persistence Task (MTPT-C). Importantly, behavioral and self-report measures of distress tolerance tend not to correlate strongly with one another (McHugh et al., 2011), suggesting that the two method types capture related but distinct aspects of the construct. A behavioral persistence task like the PASAT-c should not be treated as interchangeable with a self-report scale. McHugh and colleagues also documented a ceiling effect on the standard PASAT-c — many participants reaching the maximum — which motivated newer titrated/adjusting versions designed to spread scores more evenly across participants.
The PASAT-c Research Platform is provided as a free resource for the scientific community to facilitate behavioral research. By using this platform, you agree to the following terms.
This platform is intended exclusively for academic research, clinical assessment, and educational purposes. Use for commercial purposes — including, but not limited to, employee screening, consumer assessment, or market research — is strictly prohibited without prior written authorization.
Researchers using this platform are required to provide appropriate citations in all publications, presentations, and reports that include data collected using this system.
If you modify the task parameters or procedures from the standard PASAT-c protocol, you must clearly describe these modifications in your methodology section. Substantial deviations from the original paradigm should be noted when interpreting results in the context of existing literature.
As the researcher, you are responsible for ensuring compliance with all applicable data-protection regulations (including GDPR, HIPAA, and others) in your jurisdiction:
Researchers must obtain approval from their Institutional Review Board (IRB) or Ethics Committee before collecting data from human participants. This platform does not provide IRB approval or substitute for institutional oversight. Because the PASAT-c is designed to induce distress, particular care should be taken with consent, debriefing, and the wellbeing of participants.
You retain full ownership of all data collected through this platform. We do not access or analyze your research data.