Science

Lay's GPS scale: a complete analytical report on action control

How to read this article.This is a complete reference to the General Procrastination Scale (GPS), which 40 years later remains the number one tool for measuring procrastination as an enduring personality trait. The text brings together peer-reviewed sources, historical context, and practical insights for product application.

1. Author and context of creation

The author of the scale is a Canadian psychologistClarry H. Layfrom the Department of Psychology at York University (Toronto). The instrument was first described in an article by Lay, C. H. (1986). “At last, my research article on procrastination” inJournal of Research in Personality, 20(4), 474–495.

With the title of the article - “Finally, my research article on procrastination” - Lay played on the subject of his own research. Until 1986, procrastination was studied almost exclusively as a situational phenomenon—students delaying their studies. Lay set a new challenge: measure procrastination asstable personality trait (trait), manifested not only in studies, but also in everyday life - correspondence, paying bills, packing for travel, buying gifts.

2. How the scale was created: step by step

The development was carried out according to the recommendations for the construction of personality questionnaires proposed by Jackson (Jackson, D. N., 1970) - at that time it was the standard methodology of psychometrics: a wide pool of statements → testing on a sample → statistical selection of the best items → final short version.

Late 1970s - early 1980s

Statement of the problem

Lay decides to measure not situational procrastination (“I didn’t turn in my coursework on time”), but a general disposition—the tendency to put things off regardless of the context: studying, correspondence, bills, preparing for a trip, buying gifts.

Stage 1 - draft pool

≈36 statements in true/false format

A wide pool of statements is collected: 18 “direct” (describe procrastination directly—“I often procrastinate”) and 18 “reverse” (describe punctuality—“I usually get down to business right away”). The binary format at the start is a standard technique for the initial collection of points.

Stage 2 - factor analysis

Selection up to 20 items

The pool is tested on a sample, and statements are statistically selected that “load” most heavily on a single general factor—that is, they are most closely related to the answers to the remaining items. Weak, ambiguous or redundant items are eliminated.

Stage 3 - change the response format

From “true/false” to 5-point Likert scale

The final format is a 5-point scale (“completely uncharacteristic” → “completely characteristic”). This provides more statistical information per item than a binary response and more accurately captures intermediate degrees of trait expression rather than just extremes.

1986

Publication

The final version - 20 statements (10 forward + 10 reverse) - is published along with three validation studies inJournal of Research in Personality.

🔄
Why are “reverse” points needed?This is protection against acquiescence bias - the tendency of some people to automatically agree with almost everything they are asked to read, without really thinking about it. If all 20 statements expressed procrastination directly, the “agreeing” respondent would receive an inflated score simply for the manner of answering, and not for the actual trait. When half of the points are formulated in the opposite direction, “mechanical agreement” with everything cancels out the effect and does not distort the final score.

3. Three studies in the original 1986 article

Lay's article is not just a description of the questionnaire, but a program of several studies demonstrating its validity. Until then, personality questionnaires were most often tested by comparing one self-report with another self-report—essentially a vicious circle. Lay was the first to systematically link the self-report GPS scale with objectively observable behavior of people - this is calledcriterion validity, and it was she who transformed GPS from “just another questionnaire” into a tool that became the standard for decades.

📮
Study I - Postal Return.Participants were given a questionnaire and asked to return it by mail by a certain deadline. The researcher checked the GPS score against the date of the postmark on the envelope—that is, he did not ask people whether they were late, but recorded the fact of the delay objectively, according to an independent source. People with higher GPS scores were statistically significantly more likely to return the questionnaire late.
🗂️
Study II - analysis of personal projects (Little's technique).Participants 1–3 weeks after the first test were asked to fill out the Personal Projects Analysis method. Procrastinators assessed their current projects differently than non-procrastinators: higher perceived stress and lower enjoyment of the process of working on them - that is, a high GPS score was accompanied not just by “laziness,” but by literally a different subjective experience of their own affairs.
✈️
Study III—Envelope Experiment.Another behavioral test: As part of a postal service efficiency study, participants were given envelopes asking them to post them on a certain date after receiving them. Accuracy in keeping the date was then compared to the GPS score - again a behavior rather than a self-image.

Some retellings ascribe to Lay's paper the Test-Wait-Test-Act process model of procrastination, based on Miller, Galanter, & Pribram's (1960) classic work on the structure of goal-directed behavior—an idea consistent with the general spirit of the cognitive psychology of the era, but it is more illustrative and not central to the 1986 methodology.

4. Psychometric characteristics - and what they mean in practice

Psychometrics without explanation looks like a set of mysterious numbers. Below is what each number literally means.

🧩
Cronbach's α is internal consistency.Shows how consistently different test items behave with each other. If all 20 items really measure the same thing—procrastination—a person who answers “that's about me” to one item should be more likely to respond similarly to the others. α = 0.82 indicates high consistency: items “pull in one direction” rather than measuring random noise or different, unrelated things. Conventional psychometric benchmarks: <0.6 - weak, 0.7-0.8 - acceptable, 0.8-0.9 - good, >0.9 - sometimes means that items simply duplicate each other.
🔁
Test-retest reliability—stability over time.The same people are tested twice at intervals (for example, several weeks or months) and see how consistent the results are. r ≈ 0.80 means: a person with a high score today is likely to maintain a high score later. This confirms that GPS is measuring an enduring characteristic of personality, rather than momentary mood, workload this week, or fatigue on the day of filling out the questionnaire.
Cronbach's α0,82–0,90
Test-retest reliability (r)≈0,80
Characteristic Meaning Note
Cronbach's α ≈ 0,82–0,90 In the original ≈0.82; in independent replications, stable 0.80–0.90
Test-retest reliability r ≈ 0,80 Ferrari, 1989; shows the stability of a trait over time
Translations and adaptations Dozens of languages Spanish, Italian, Turkish, Greek, Portuguese, Chinese, German, Hindi, etc.
Factor structure Single-factor (Lei) vs multi-factor (replication) Turkish, Italian, Greek - two-factor; Hasanagić & Özsağır (2018, N=480) - five-factor
Discussion about factor structure.Lay positioned the scale as single-factor, with “procrastination” as a single dimension. But a number of confirmatory factor analyzes in different cultures have found multifactorial solutions: for example, “quality of planning” is separate from “procrastination” as such. Even when several factors are identified, researchers usually continue to use the total score for the sake of comparability with the body of previous work, which always considered the total score.

5. Family of related scales

Tool Author, year Items Peculiarity
GPS Lay, 1986 20 Basic one-factor model; 10 forward + 10 reverse
GPS-9 Sirois, Yang & van Eerde, 2019 9 Compact version; α=0.88–0.91 on samples N=4492 on average
AIP (Adult Inventory of Procrastination) McCown & Johnson, 1989 15 Chronic behavioral procrastination in adults
PPS (Pure Procrastination Scale) Steel, 2010 12 Factorial selection from three scales; purified irrational delay index
GPS-Lodha Lodha et al., 2016 23 Four domains (academic, work, medical, civil); range 23–115
📊 Key fact about GPS vs AIP.Ferrari (1992) showed that both scales were significantly correlated with actual behavioral inhibition but were NOT correlated with each other. This suggests that they are measuring different motivational mechanisms: GPS is more likely to capture avoidance, while AIP is more likely to capture arousal/thrill seeking from a deadline.
What does this mean in practice?There is no one “correct” tool for measuring procrastination—there is a family of scales, each of which shifts the focus slightly. GPS is a general tendency to procrastinate as a personality trait. GPS-9 - the same feature is more compact, convenient for long batteries of questionnaires, where saving the respondent’s time is important. AIP is specifically a chronic behavioral procrastination in adults outside of an educational context. PPS/IPS are indices of irrational delay “cleared” of related constructs (punctuality, organization). GPS-Lodha - broken down by life domain (work separately from health, for example). Practical conclusion for the product: fast screening - GPS-9; it is necessary to distinguish between motivational mechanisms (fear of failure versus the craving for adrenaline of a deadline) - a combination of GPS + AIP, one point is not enough; domain specificity is important - look towards GPS-Lodha.

6. Connection to Personality (Big Five)

🧬
What is the Big Five?This is the dominant model of personality structure in modern psychology: five relatively independent dimensions—openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism (emotional instability). Each dimension is a continuum, and every person occupies some point on it. The connection of procrastination with these five axes shows which part of the personality is “responsible” for the tendency to procrastinate and which is not.

Meta-analysis Steel (2007,Psychological Bulletin) on ~691 correlations from dozens of works revealed stable connections:

Conscientiousnessr ≈ −0,55…−0,75
Neuroticismr ≈ +0,20…+0,35

Conscientiousness: The most repeatable outcome in the entire field.This dimension combines several facets - organization, reliability/sense of duty, desire for achievement, self-discipline, caution in decisions. Not all of them “hit” procrastination in the same way: they are most strongly connectedself-discipline(the ability to finish what you start without getting distracted) anddesire for achievements(how much a person is generally driven by a goal for which it is worth hurrying). This is logical: procrastination is primarily a failure of self-regulation, and not a lack of ability, and self-discipline is a facet of personality that is specifically about the mechanics of “pull yourself together”, and not about the content of the task.

Neuroticism: moderate relationship through anxiety.People high in neuroticism worry more about the possibility of failure, criticism, or the consequences of making a mistake. This anxiety in itself is unpleasant, and the natural reaction is to move the source of anxiety further in time and not think about the task. Hence the moderate (not strong) positive relationship: anxiety fuels avoidance, but is not its sole cause.

Extraversion, agreeableness, openness to experience: weak and not always reproducible connections.This is expected: these three dimensions do not describe the mechanisms of self-regulation, but the content of personality - sociability, kindness towards people, curiosity about new experiences. None of them is directly responsible for the ability to force yourself to do an unpleasant task in a timely manner, so the connection with procrastination is weak and unstable from sample to sample.

Persistence (Grit) and impulsiveness.In addition to the Big Five, a negative relationship between procrastination and indicators of persistence (Grit - the ability to maintain interest and effort for years for a long-term goal) and a positive connection with impulsivity (the tendency to act under the influence of a momentary impulse without assessing the consequences) is consistently reproduced. Both results logically continue the story with conscientiousness: procrastination is tightly connected precisely with the system of self-regulation of behavior over time, and not with intelligence, the nature of communication with people, or the breadth of interests.

7. Genetics: what the twin study showed

👯
How is the “heritability” of a character trait measured?The logic of twin studies: identical (monozygotic) twins have almost 100% of their genes in common, fraternal (dizygotic) twins have an average of 50%, like ordinary brothers and sisters. If identical twins are significantly more similar to each other than fraternal twins for some trait, the difference in similarity is statistically attributed to genetics. This gives an estimate of heritability as a percentage - not the proportion “fixed” in the genes of a particular person, but the proportion of differences between people in the population that is statistically explained by genetic differences.

First Twin Study on Procrastination -Gustavson, Miyake, Hewitt & Friedman (2014, Psychological Science). With this sample, it was possible for the first time to separate the genetic and environmental components of the trait.

46%
Heritability (genetics) - ≈46%
Environment and individual experience - ≈54%
Key Findings:
  • Heritability of procrastination -≈46%; heritability of impulsivity is ≈49%.
  • At the phenotypic level (that is, according to observed behavior), they are separable - they correlate with each other at the level of r≈0.65, which means that they are not the same thing. But at the genetic level they are practically inseparable: genetic correlation ≈1.0. This means that the same set of genetic factors underlies both traits—procrastination and impulsivity seem to “grow from the same root,” even if they look different on the outside.
  • Near72%Common genetic variation of procrastination is shared with the ability to manage goals in everyday situations (goal-management ability) - a general executive function of the brain responsible for maintaining a goal and inhibiting distracting impulses.

What does this mean meaningfully?Procrastination, apparently, is not a separate, isolated “breakdown”, but rather one of the visible symptoms of a more general, genetically determined ability to self-manage goals. This is consistent with what is seen at the personal level (section 6): the center of the story is not laziness or character, but a mechanism of self-regulation. However, a heritability of ≈46% is not a death sentence at the individual level: it describes how differences are distributedbetween people in a population, and not that someone’s specific procrastination is “half in the genes” and therefore cannot be worked on.

8. Neurobiology

Study of the volume of the amygdala. Schlüter, Pinnow, Güntürkün & Genç (2018, Psychological Science) on a sample of 264 people, using MRI, they measured the volume of individual brain areas and the functional connectivity between them - that is, how consistently two areas of the brain are activated over time, which indirectly indicates the strength of “communication” between them. At the same time, the self-reported level of action control was assessed.

🧠Amygdala
larger in volume

weakened connection
🎯 dACC
brake control
🔬
What do these two areas of the brain do?The amygdala (amygdala) is a structure that is one of the first to react to a potential threat or negative consequences, including socio-psychological ones (“what if I fail”) - it generates anxiety. The dorsal anterior cingulate cortex (dACC) is involved in monitoring conflicts between different drives and in the transition from “thinking” to “acting,” especially when action requires effort. The connection between them is, roughly speaking, a channel through which the alarm signal can be “taken under control” and not interfere with getting to the point.

The result: People with poor action control have larger amygdala and weaker functional connectivity between the amygdala and the dACC. Author's interpretation: People with larger amygdala worry more about the negative consequences of an action, which encourages them to hesitate; and a weakened amygdala-dACC connection makes it difficult to effectively suppress this anxiety and move on to action despite it.

The Zeigarnik effect and “closing the loop.”The classic effect (Zeigarnik, 1927): unfinished tasks are remembered better than completed ones and create background cognitive tension - the brain seems to keep an “open tab” on the task until it is closed in one way or another. Significant work - Masicampo & Baumeister (2011,Journal of Personality and Social Psychology): Making a specific plan for an unfinished goal removes its distracting cognitive effect almost as effectively as actually completing the task. The brain “postpones” the task in the background - but only if it already has a clear plan: when exactly, where exactly, how it will be done.

🧠 Practical conclusion for products:Instead of simply “save for later,” it is more useful to force the user to record a specific “when exactly” - this is closer to the real mechanics of the Zeigarnik effect than arbitrary formulations about “motivation.” A vague “I’ll do it later” doesn’t close the open loop and continues to be a background distraction; a specific plan with a launch date and condition - closes.

9. Temporal Motivation Theory

Influential work -Steel & König (2006, Academy of Management Review). The formula combines expectancy theory (Vroom), cumulative prospect theory (Kahneman and Tversky), needs theory (Maslow, Murray) and hyperbolic discounting (Ainslie, Laibson) into a single model of how motivation to act right now is formed:

Motivation = (Expectancy × Value) / (1 + Impulsiveness × Delay)

Let's look at each variable separately - in simple words:

VariableWhat does it mean
Expectancy (E)Subjective probability of success is how much a person believes that if he takes on a task, he will succeed. Low confidence in success reduces motivation even when the value of the result is high.
Value (V)The subjective value of the result is how important or enjoyable the task is for this person personally. Not objective importance, but a personal feeling of “why do I need this.”
Impulsiveness (I)An individual trait is how strongly a person devalues ​​delayed versus immediate rewards. High impulsivity—a strong preference for “now” over “later.”
Delay (D)Time until a deadline or until a reward/consequence is received. The longer the delay, the more the motivation “erodes” - this is hyperbolic discounting.

The denominator “1 + Impulsiveness × Delay” is a mathematical form of hyperbolic discounting: the longer the delay (Delay) and the higher the personal impulsivity, the more the final motivation falls, and the decline is non-linear - it accelerates sharply as the delay increases, rather than going evenly. The unit in the denominator is important: without it, at Delay → 0 (right before the deadline), the formula would give endless motivation, which is physically meaningless.

Far from deadline Deadline Motivation increased motivation accelerates as reduce latency

Shape of the TMT curve: motivation remains low most of the time and increases sharply as the delay decreases—hence the subjective feeling of “I only turn on when there’s a deadline.”

Calculation parameterMeaning
Expectancy (E)0,8
Value (V)8 out of 10
Impulsiveness (I)3 (conventional unit of trait)
Delay = 30 days → Motivation≈ 0,07
Delay = 1 day → Motivation≈ 1,60

This is an illustrative calculation with conditional parameters (not data from a specific study): for the same person and the same task, the formula predicts an increase in motivation by almost 23 times when the delay is reduced from 30 days to 1 day - this is exactly what “I turn on only at the last moment” looks like mathematically.

An Extensive Empirical Application of TMT to Procrastination—A Meta-Analytic ReviewSteel (2007, Psychological Bulletin, ~691 correlations from dozens of studies). This work remains the most cited review in the field to this day, and it was the work that brought together the data on personality (Section 6), genetics, and the overall theoretical framework of TMT into one picture.

10. Active vs passive procrastination

Real and frequently cited work:Chu & Choi (2005, The Journal of Social Psychology). The authors divided procrastinators into two types with formally the same “result”—both delay starting work—but fundamentally different internal mechanics.

🌫️

Passive type - “tourist in the fog”

  • Postpones not out of decision, but out of paralysis of indecision
  • Poor sense of control over time
  • Lower self-efficacy - does not believe that he will have time to cope
  • Avoidant coping strategies: not thinking about a task until it becomes very critical
  • Results and performance suffer noticeably
🌊

Active type - “surfer waiting for a wave”

  • Consciously procrastinates - prefers to work under deadline pressure
  • Maintains a high sense of control over the process
  • Self-efficacy at the level of people without procrastination at all
  • Problem-oriented coping strategies: a focused breakthrough in a short time
  • Results and performance are close to those of “non-procrastinators”

With a comparable degree of formal “procrastination,” active procrastinators are closer to people without procrastination than to passive procrastinators in terms of time use, sense of control, self-efficacy, coping strategies, and even academic performance. The difference is notWhatperson does (both start late), but in thatWhyAndwith what internal state.

Practical consequence.The raw GPS score does not differentiate between active and passive types, so two groups of users with the same score may need contrasting interventions: the passive type needs to break down the task into small, non-critical steps and reduce pressure; the active type, on the contrary, may need to maintain a manageable tension of the deadline - aggressive “simplification” and removal of urgency sometimes reduces its effectiveness.

11. Planning Fallacy and underestimation of time

People systematically underestimate the time it takes to complete tasks, even based on past experience with similar tasks. The term and initial experiments are associated with the work of Kahneman & Tversky in the late 1970s, with further empirical development in Buehler, Griffin & Ross (1994) and subsequent work by this group.

🔭
Mechanism: “inside view” versus “outside view.”Kahneman described two ways to predict the duration of a task. “Inside view” - imagine exactly how this particular task will be performed, step by step, under ideal conditions without failures; This is a natural way of thinking, and it is the one that produces systematically low estimates, because in our minds we almost always imagine a smooth scenario. “Outside view” - instead of a detailed scenario, look at statistics: how much time, on average, similar tasks actually took in the past, regardless of the details of the current one. The practice of reference class forecasting (Flyvbjerg) - the deliberate use of an external view - reduces systematic underestimation in large projects more accurately than trying to “think everything through in advance” even more carefully.

The general direction of the effect is consistently reproduced in cognitive psychology on a variety of types of tasks - from student coursework to the construction of infrastructure facilities.

Practical conclusion for the bot:It is more useful to calculate the individual time estimation distortion coefficient empirically - by comparing the user's stated estimate with the actual execution time from the logs of that particular person, rather than postulating a fixed percentage from the literature. Over time, the accumulated history of “promised → did” turns into a personal analogue of the “look from the outside”: the system can suggest not an abstract percentage, but a specific amendment based on the past behavior of this particular user.

12. Procrastination and health

Sirois (2015, Journal of Behavioral Medicine) in a sample of people diagnosed with hypertension/cardiovascular disease (182 diagnosed people versus 564 controls) showed that trait procrastination was significantly associated with the presence of these diseases, even after adjusting for age, race, education, and other personality traits.

😮‍💨 Chronic procrastination
⚡ Chronic stress
🫀 Psychophysiological load
🩺 Risk of hypertension/CVD

The cycle closes: stress and avoidance reduce the resource for timely seeking help, which increases risks and fuels further procrastination.

Mechanism.Chronic stress from constant procrastination is not an abstract metaphor, but a measurable psychophysiological burden: repeated worry about what has not been done keeps the body in a state of hypermobilization longer than necessary, and this over time increases vulnerability to cardiovascular problems. This is a stable and actively developing research line - Sirois is systematically studying the other side of the same mechanism.

The role of self-compassion.A separate line of work by Sirois (including the article “Procrastination and stress: Exploring the role of self-compassion” cited below) shows that people who are able not to punish themselves for the fact of procrastination, but to treat themselves with understanding, break the vicious circle more effectively - self-flagellation after a breakdown adds another layer of stress on top of the original one, while self-compassion reduces this secondary load and makes it easier to return to the task. Sirois also described the connection between procrastination and delay in seeking medical help, another mechanism through which procrastination directly affects health: the same quality that prevents you from sitting down for a report on time, prevents you from making an appointment with the doctor on time.

13. The Economics of Procrastination

Pension savings.The classic line of research is Choi, Laibson, Madrian & Metrick (several papers from 2003–2004, including “For Better or For Worse: Default Effects and 401(k) Savings Behavior”). Procrastination causes people to delay both the decision to increase contributions and the decision to leave a plan during automatic enrollment.

⚙️
How “default” turns procrastination to its advantage.If by default an employee does NOT participate in the pension plan, procrastination means “I’ll never get around to joining”—lost benefits. But if by default the employee ALREADY participates (automatic/default enrollment), and in order to exit, you need to take an active action - the same tendency to postpone the decision now works in favor of savings: the person simply does not get around to unsubscribing. This is one of the most cited examples of “choice architecture” in behavioral economics: don’t fight the propensity to procrastinate, but design the system so that inaction leads to a good outcome rather than a bad one.
≈26.3 million ₽
Start at 25
(40 years of contributions)
≈12.2 million ₽
Start at 35
(30 years of contributions)

Illustrative calculation (not research data): contribution 10,000 ₽/month, yield 7% per annum, retirement at 65 years old. A ten-year delay in the start almost halves the final amount - although the difference in the money contributed is much less than the difference in the end. The main effect is the time lost on compound interest, not the missed payments themselves.

This shows that sometimes the correct use of the very nature of procrastination - through the architecture of choice, and not through attempts to “re-educate” a person - works more effectively than any appeal to willpower.

14. Historical background: “thief of time”

Phrase«Procrastination is the thief of time»belongs to an English poet and Anglican priestEdward Young (1683–1765)and comes from his poem"Night Thoughts", published in parts in 1742–1745 (exact line - book 1, line 393).

"Night Thoughts" is a long meditative poem in blank verse on the themes of death, the immortality of the soul and the transience of time, written largely under the influence of Young's loss of loved ones. In the 18th–19th centuries, the poem became one of the most read and translated works of English literature of its time, giving rise to an entire genre of “graveyard poetry” that preceded romanticism. Half a century after its publication, from 1795 to 1797, the poet and artist William Blake created a series of several hundred watercolor illustrations of “Night Thoughts”—in itself a testament to how firmly the poem had become embedded in culture.

The line about the “thief of time” has outlived both the author and the poem itself as an independent work: today almost no one reads Young in its entirety, but the phrase continues to live separately - as one of the most recognizable images of procrastination in the English language, almost three centuries after its publication.

15. Criticisms and limitations of the tool

  • Self-report.Subject to social desirability and inaccurate self-reflection - a standard limitation of any personality questionnaire.
  • Criterion contamination.Some items measure related things—general punctuality, organization, being busy—rather than irrational delay per se. This was directly recognized as the reason for the appearance of the “purified” PPS and IPS scales.
  • Weak relationship with objective performance.In Lay's original study, the association between GPS and test scores was close to zero. In highly structured, time-constrained situations, procrastinators perform much the same as others; the difference manifests itself in semi-structured, self-organizing tasks. This is one of the most underrated parts of the scale's history.
  • Incompatible with AIP.Ferrari (1992) showed that GPS and AIP were significantly correlated with real-world latency, but were not correlated with each other—a recommendation to use a combination rather than a single tool.

Conclusion

Lay's GPS is a reliable and well-studied indicator of dispositional procrastination with a nearly 40-year replication history. Over four decades, the scale has gone from 36 draft statements in a binary format to 20 refined items, the psychometrics of which have been replicated across dozens of cultures and languages. A large body of related, real-life research has accumulated around it - from twin studies of genetics to neuroimaging of the amygdala and dACC, from the behavioral economics of pension savings to the psychophysiology of cardiovascular risk.

Putting the whole picture together, it is clear that procrastination is not a single fact of laziness or character flaw, but the tip of several intersecting systems: a roughly half-genetic ability to manage goals and inhibit distracting impulses (Section 7); the personal axis of self-regulation, where the decisive role is played not by character in general, but specifically by self-discipline and resistance to anxiety (section 6); a neural circuit in which the amygdala generates anxiety about consequences and the dACC is either successful or not so successful in translating intention into action (section 8); and the mathematical principle of hyperbolic discounting, which causes almost anyone's motivation to skyrocket as soon as a deadline approaches (Section 9).

Practical conclusions for those who build a product around this topic: firstly, GPS works well as a self-reported disposition metric, but its scores should be supplemented with behavioral data - real “promised → delivered” logs, and not just a questionnaire. Secondly, the scale does not distinguish between active and passive types of procrastinators (section 10), so the same score may correspond to users who need interventions that are opposite in spirit - for some, to reduce the pressure and break the task into steps, and for others, on the contrary, to maintain a manageable tension of the deadline. Third, where it is almost impossible to overcome the tendency to procrastinate through willpower, it is often more effective to change the architecture of choice - as automatic pension enrollment programs have done (Section 13) - so that inaction itself leads to the desired result. And finally, fourthly: a specific plan with a date and trigger condition closes the background cognitive stress of the task (Zeigarnik effect, section 8) much more effectively than a vague “I’ll do it later” - and this is perhaps the simplest and at the same time the most evidence-based advice in the entire 40-year history of research on the scale.