Working Memory
Recent Probes Task
Also called: recent-probes task, recent negatives task, item recognition with recent probes, proactive interference task
Measure proactive interference in short-term memory: four letters, a short wait, then a letter to judge, scored as the slowing when the letter was in the previous set.
10-11 minTimed: desktop or laptop with a mouse or trackpad
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Preview any test free after you sign up.
- Papers, 2021 to 2025
- 17
- Evidence grade
- D
- Registered clinical studies
- 1
What it measures
Measures interference from the previous trial in short-term memory. Participants see four letters, wait 3 seconds, and decide whether a capital letter was in the set. The main score is how much slower they reject a letter that was in the previous set.
Constructs
- Proactive interference
- Working memory
- Interference resolution
Research uses
- Working memory and cognitive control research
- Aging research
- EEG and brain imaging studies of interference
How the paradigm works
The Recent Probes Task measures proactive interference in short-term memory: how much a letter seen a moment ago gets in the way of a new decision. On each trial four small letters appear around a cross for 1,500 ms. The cross then stays alone for 3,000 ms, and a capital letter, the probe, stays until the participant answers or 3,000 ms pass. Participants press one key if the probe was one of the four letters just shown and the other if it was not, as quickly and accurately as they can.
More about the paradigm
The task follows Monsell (1978) in the letter version of Jonides and colleagues (1998), with the four probe types of D'Esposito and colleagues (1999) and the timing that Thompson-Schill and colleagues (2002) describe. Half the probes were in the set and half were not. Half of each kind were also in the previous trial's set (recent) and half were in neither of the last two sets. A recent probe that is not in the set, a recent negative, feels familiar, so saying no to it takes longer and goes wrong more often. Every set keeps two letters of the set before it, so a repeated letter never tells the answer. The letters are 12 consonants; the set is shown in small letters and the probe as a capital, so the probe is matched by letter and not by shape. Each session's order comes from its seed, two unscored trials open the test, and the F and J keys for yes and no are counterbalanced across sessions. Six practice trials give feedback; the test gives none. The panels switch on screen frames, and each display's measured duration is saved on every trial.
The primary score is the interference effect: the median correct response time to recent negatives minus that to probes in neither of the last two sets. In young adults it is typically 50 to 100 ms (Berman, Jonides and Lewis 2009). Scores also give the same difference in error rates, the effect of recency on probes in the set, the median correct response time and accuracy for each probe type, and anticipations (responses under 100 ms, left out of the response times). The effect is meant for group comparisons rather than for one person's score: its reliability for individuals is low (Snijder and colleagues 2024). Scores support research and do not diagnose.
Scores
- Interference Effectms
- Median correct response time to recent negatives (probes in the previous set but not this one) minus that to probes in neither of the last two sets.
- Positive when a recent letter slows the no answer; smaller means less interference
- Interference Effect, Errors%
- Error rate on recent negatives minus the error rate on probes in neither of the last two sets.
- Positive when a recent letter causes more errors
- Recency Effect, Probe in the Setms
- Median correct response time to probes in this and the previous set minus that to probes in this set only.
- Negative when recency speeds the yes answer
- Median RT, Recent Negativesms
- Median correct response time to probes in the previous set but not this one.
- Lower is faster
- Median RT, Other Negativesms
- Median correct response time to probes in neither of the last two sets.
- Lower is faster
- Median RT, Recent Positivesms
- Median correct response time to probes in this and the previous set.
- Lower is faster
- Median RT, Other Positivesms
- Median correct response time to probes in this set only.
- Lower is faster
- Accuracy, Recent Negatives%
- Share of probes in the previous set but not this one answered correctly.
- Higher is better
- Accuracy, Other Negatives%
- Share of probes in neither of the last two sets answered correctly.
- Higher is better
- Accuracy, Recent Positives%
- Share of probes in this and the previous set answered correctly.
- Higher is better
- Accuracy, Other Positives%
- Share of probes in this set only answered correctly.
- Higher is better
- Overall Accuracy%
- Share of scored trials answered correctly. A missed response is incorrect.
- Under 80% is flagged for review
- Fewest Correct Negativestrials
- Correct response times for the kind of probe not in the set with fewer. The main score is a difference of their medians.
- Fewer than 10 is flagged for review
- Anticipationscount
- Responses under 100 ms, left out of the response times.
- Context for the scores
- Missed Responsescount
- Scored trials with no response within 3,000 ms.
- Context for the scores
Use and evidence
17 papers from 2021 to 2025 name this task in their title or abstract, down 32% compared with 2016 to 2020. That ranks it 341 of 395 test families in our census. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 3 |
| 2024 | 2 |
| 2023 | 5 |
| 2022 | 3 |
| 2021 | 4 |
| 2020 | 7 |
| 2019 | 4 |
| 2018 | 4 |
| 2017 | 5 |
| 2016 | 5 |
| 2015 | 4 |
| 2014 | 6 |
| 2013 | 5 |
| 2012 | 1 |
| 2011 | 5 |
| 2010 | 3 |
| 2009 | 4 |
| 2008 | 3 |
| 2007 | 1 |
| 2006 | 2 |
| 2005 | 1 |
| 2004 | 2 |
| 2003 | 1 |
| 2002 | 0 |
| 2001 | 0 |
| 2000 | 2 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade D
Little published measurement evidence found.
- Papers on test-retest reliability
- 0
- Papers on reliability or validity
- 2
- Registered clinical studies using it as an outcome
- 1
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on this task, not ConductCognition's version of it.
Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Trial
- An empty panel for 500 ms, a cross for 500 ms, four small letters around the cross for 1,500 ms, the cross alone for 3,000 ms, then a capital letter until a response or 3,000 ms.
- Letters
- 12 consonants: b, d, f, g, h, l, m, n, p, q, r and t. Each set keeps 2 letters of the set before it and adds 2, so a repeated letter never tells the answer.
- Probe types
- 20 trials of each of 4 types, 80 in all: in this set and the previous one, in this set only, in the previous set only (recent negative), and in neither of the last two sets.
- Trial order
- A seeded random order, saved with the trial data, with no more than 3 probes of one answer in a row. 2 unscored trials open the test, so every scored trial has two before it.
- Keys
- F and J for in the set and not in the set, counterbalanced across sessions.
- Practice
- 6 trials with feedback. The test gives none.
- Timing
- The panels switch on screen frames, and each display's measured duration is saved on every trial.
- Response times
- Correct responses of 100 ms or more enter the medians.
The interference score is meant for group comparisons rather than one person's score, because its reliability for individuals is low (Snijder and colleagues 2024).
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: recent_probes_test from the scored run, less the two opening trials, F or J for a capital letter in this trial's set or not (chance 50%, half the probes are in the set) within 3,000 ms, correctness from the key for the probe. Below chance invalidates and not above chance is flagged. One key on 90% of answers invalidates when accuracy is not above chance. Missed responses and responses under 100 ms follow the timed-test rule: more than max(3, 25%) missed or more than max(2, 20%) under 100 ms invalidate, and any is flagged. Completing fewer than half the planned trials invalidates, and any other incomplete run is flagged. Accuracy under 80%, or fewer than 10 correct responses to either kind of probe not in the set, is flagged for review.
Why these rules
Accuracy significantly below chance (one-sided exact test, p < .05) cannot come from guessing; it means the response mapping was reversed or the task was not followed (Slick, Sherman and Iverson 1999 treat below-chance forced-choice performance as definite response bias). Chance-level accuracy can be a real result in impaired participants, so it is only flagged. One option on at least 90% of answered trials, when the correct options are balanced and accuracy is not above chance, means the answers did not depend on the items (the task counterpart of long-string responding in surveys; Curran 2016). Missed responses and responses under 100 ms keep the platform's timed-test rule: more than max(3, 25%) missed responses or more than max(2, 20%) responses under 100 ms invalidate, and any is flagged. A response under 100 ms cannot be a reaction to the stimulus. Young adults answer more than 92% of every probe type correctly in this task (Berman, Jonides and Lewis 2009, https://doi.org/10.1037/a0014873), so accuracy under 80% means the response times may not reflect memory for the set. The main score is a difference of two medians of correct response times, so it needs enough of each. A hidden tab lets the study, delay or response timers run without the participant, so a percentile is withheld after one.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Trials of each probe type | 20 | 12 to 32 |
| Wait before the probe (ms) | 3,000 | 1,000 to 4,000 |
Yearly plans can change settings within these limits. Free and pay-as-you-go studies use the standard settings.
Need a setting this test does not have?
Devices and languages
- Timed test: participants need a desktop or laptop with a mouse or trackpad. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- No percentile. Scores and quality flags only. How norms work
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Recent Probes Task in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Monsell, 1978; Jonides et al., 1998), with the standard settings (trials of each probe type: 20; wait before the probe (ms): 3,000). The main score was interference effect (median correct response time to recent negatives (probes in the previous set but not this one) minus that to probes in neither of the last two sets).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Monsell, S. (1978). Recency, immediate recognition memory, and reaction time. Cognitive Psychology, 10(4), 465-501. https://doi.org/10.1016/0010-0285(78)90008-7
- Jonides, J., Smith, E. E., Marshuetz, C., Koeppe, R. A., & Reuter-Lorenz, P. A. (1998). Inhibition in verbal working memory revealed by brain activation. Proceedings of the National Academy of Sciences, 95(14), 8410-8413. https://doi.org/10.1073/pnas.95.14.8410The letter version with recent and not recent probes.
- D'Esposito, M., Postle, B. R., Jonides, J., & Smith, E. E. (1999). The neural substrate and temporal dynamics of interference effects in working memory as revealed by event-related functional MRI. Proceedings of the National Academy of Sciences, 96(13), 7514-7519. https://doi.org/10.1073/pnas.96.13.7514Recent probes in and not in the set.
- Thompson-Schill, S. L., Jonides, J., Marshuetz, C., Smith, E. E., D'Esposito, M., Kan, I. P., Knight, R. T., & Swick, D. (2002). Effects of frontal lobe damage on interference effects in working memory. Cognitive, Affective, & Behavioral Neuroscience, 2(2), 109-120. https://doi.org/10.3758/CABN.2.2.109The timing, the four probe types and two letters kept from each set.
- Jonides, J., & Nee, D. E. (2006). Brain mechanisms of proactive interference in working memory. Neuroscience, 139(1), 181-193. https://doi.org/10.1016/j.neuroscience.2005.06.042Review of the task.
- Berman, M. G., Jonides, J., & Lewis, R. L. (2009). In search of decay in verbal short-term memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(2), 317-333. https://doi.org/10.1037/a0014873Size of the effect and accuracy by probe type.
- Snijder, J.-P., Tang, R., Bugg, J. M., Conway, A. R. A., & Braver, T. S. (2024). On the psychometric evaluation of cognitive control tasks: An investigation with the Dual Mechanisms of Cognitive Control (DMCC) battery. Behavior Research Methods, 56(3), 1604-1639. https://doi.org/10.3758/s13428-023-02111-7Reliability of the recency effect.
Source papers
- Monsell S (1978). Recency, immediate recognition memory, and reaction time. Cognitive Psychology, 10(4):465-501.Paradigm source, Adults
- Berman MG, Jonides J, Lewis RL (2009). In search of decay in verbal short-term memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(2):317-333.Paradigm source, Young adults
Questions
How long does the Recent Probes Task take?
About 10 to 11 minutes, as listed in the test library.
What device does a participant need?
Timed test: participants need a desktop or laptop with a mouse or trackpad. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Interference Effect, Interference Effect, Errors, Recency Effect, Probe in the Set, Median RT, Recent Negatives, Median RT, Other Negatives, Median RT, Recent Positives, Median RT, Other Positives, Accuracy, Recent Negatives, Accuracy, Other Negatives, Accuracy, Recent Positives, Accuracy, Other Positives, Overall Accuracy, Fewest Correct Negatives, Anticipations and Missed Responses. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Recent Probes Task used in research?
17 papers from 2021 to 2025 name it in their title or abstract, down 32% compared with 2016 to 2020. That ranks it 341 of 395 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.14.0, counted 31 October 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Recent Probes Task have?
In the published research, 0 papers report its test-retest reliability and 1 registered clinical study uses it as an outcome (evidence grade D in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Trials of each probe type (12 to 32) and wait before the probe (ms) (1,000 to 4,000). Settings can be changed on yearly plans; free and pay-as-you-go studies use the standard settings.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.