Category: Tier 2 - Emotional Traits Scale: 0.0 (poor recovery) to 1.0 (excellent recovery)

Definition

Interrupt recovery measures how well users resume tasks after breaks, distractions, or context switches. It covers phone calls, notifications, tab switches, and session timeouts.

Low recovery users lose mental context and must restart from the beginning. High recovery users use environmental cues (breadcrumbs, progress bars, browser history) to continue with minimal lost progress.

Research Foundation

Primary Citation

"We found that the average time to return to a disrupted task was 23 minutes 15 seconds. Furthermore, people did not simply resume the interrupted task; rather, they engaged in an average of 2.26 intervening activities before returning to the original task." -- Mark, G., Gonzalez, V.M., & Harris, J., 2005, p. 112

Full Citation (APA 7): Mark, G., Gonzalez, V. M., & Harris, J. (2005). No task left behind? Examining the nature of fragmented work. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 321-330.

DOI: https://doi.org/10.1145/1054972.1055017

Supporting Research

"Resumption lag - the time to resume a task after an interruption - is significantly affected by the complexity of the primary task and the length of the interruption. Longer interruptions result in greater context loss and longer resumption times." -- Altmann, E.M., & Trafton, J.G., 2002, p. 41

Full Citation (APA 7): Altmann, E. M., & Trafton, J. G. (2002). Memory for goals: An activation-based model. Cognitive Science, 26(1), 39-83.

DOI: https://doi.org/10.1207/s15516709cog2601_2

Key Numerical Values

Metric	Value	Source
Average task resumption time	23 min 15 sec	Mark et al. (2005)
Intervening activities before resumption	2.26 average	Mark et al. (2005)
Resumption lag (controlled lab)	2-30 seconds	Altmann & Trafton (2002)
Error rate increase post-interruption	2x baseline	Monk et al. (2008)
Context decay half-life	15-60 seconds	Altmann & Trafton (2002)
Visual cue resumption benefit	40-60% faster recovery	Trafton et al. (2011)

Interruption Types

Type	Description	Typical Duration
`external`	Phone call, person, notification	Seconds to hours
`system`	Timeout, crash, page refresh	Instant to minutes
`self_initiated`	Tab switch, new thought, distraction	Seconds to minutes
`timeout`	Session expiration, idle disconnect	Instant

Behavioral Levels

Value	Label	Behaviors
0.0-0.2	Very Poor	Loses all context after any interruption; must restart forms from beginning; forgets goal of task after distraction; cannot recall previous steps; re-reads entire page after tab switch; session timeout causes complete task abandonment; no use of environmental cues for recovery; takes full 23+ minutes to resume complex tasks
0.2-0.4	Poor	Loses 40-60% of progress after interruption; struggles to remember where they were; re-enters data they previously completed; skips steps when resuming; high error rate post-interruption; may recognize environmental cues but doesn't effectively use them; resumes in wrong section of multi-step process
0.4-0.6	Moderate	Loses 10-30% of progress after interruption; can use breadcrumbs and progress indicators to orient; may need to review recent steps; moderate resumption lag (5-15 seconds); error rate slightly elevated after interruption; benefits from "you were here" indicators
0.6-0.8	Good	Minimal progress loss (< 10%) after interruption; quickly orients using page state, URL, form values; short resumption lag (2-5 seconds); actively seeks environmental cues; maintains mental context through moderate interruptions; can context-switch between tabs effectively
0.8-1.0	Excellent	Near-seamless recovery from interruptions; leverages all environmental cues (breadcrumbs, history, form state); < 2 second resumption lag; mental context persists through long interruptions; can resume days later using browser history; proactively creates own resumption cues (bookmarks, notes)

Trait Implementation in CBrowser

Context Loss Model

CBrowser models context decay using exponential decay modified by trait:

interface InterruptRecoveryState {
  currentTaskContext: TaskContext;
  environmentalCues: string[];      // Page elements aiding recovery
  interruptionLog: Interruption[];  // History of interruptions
  contextStrength: number;          // 0-1 memory of task context
}

interface Interruption {
  type: 'external' | 'system' | 'self_initiated' | 'timeout';
  duration: number;  // milliseconds
  timestamp: Date;
}

// Context decay during interruption
function calculateContextLoss(
  interruptRecovery: number,
  interruptionDuration: number,
  cuesAvailable: number
): number {
  const halfLife = 15000 + (interruptRecovery * 45000);  // 15-60 sec half-life
  const decayRate = Math.LN2 / halfLife;
  const baseLoss = 1 - Math.exp(-decayRate * interruptionDuration);

  // Environmental cues reduce loss
  const cueRecovery = Math.min(0.6, cuesAvailable * 0.1);

  return Math.max(0, baseLoss - cueRecovery);
}

Resumption Lag

// Time to resume after interruption
function getResumptionLag(
  interruptRecovery: number,
  contextLoss: number,
  taskComplexity: number
): number {
  const baseLag = 2000;  // 2 seconds minimum
  const complexityMultiplier = 1 + (taskComplexity * 2);  // 1x to 3x
  const recoveryFactor = 1 + ((1 - interruptRecovery) * 10);  // 1x to 11x
  const contextFactor = 1 + (contextLoss * 5);  // 1x to 6x

  return baseLag * complexityMultiplier * recoveryFactor * contextFactor;
  // Range: 2 seconds to several minutes
}

Environmental Cue Detection

// Cues that help users recover context
const environmentalCues = {
  breadcrumbs: 0.15,        // "Home > Products > Category"
  progressIndicator: 0.20,  // "Step 2 of 4"
  formValues: 0.15,         // Previously entered data visible
  pageTitle: 0.10,          // Descriptive title
  recentHistory: 0.15,      // Browser back button history
  urlPath: 0.10,            // Meaningful URL structure
  visualPosition: 0.08,     // Scroll position preserved
  notifications: 0.07       // "You have unsaved changes"
};

function calculateCueStrength(page: Page): number {
  return Object.entries(environmentalCues)
    .filter(([cue]) => page.hasCue(cue))
    .reduce((sum, [, value]) => sum + value, 0);
}

Behavior Post-Interruption

// How user behaves when resuming
function getResumptionBehavior(
  interruptRecovery: number,
  contextLoss: number
): 'continue' | 'review' | 'restart' {
  const effectiveRecovery = interruptRecovery * (1 - contextLoss);

  if (effectiveRecovery > 0.6) return 'continue';  // Pick up where left off
  if (effectiveRecovery > 0.3) return 'review';    // Review recent steps, then continue
  return 'restart';  // Begin task from start
}

Estimated Trait Correlations

Correlation estimates are derived from related research findings and theoretical models. Empirical calibration is planned (GitHub #95).

Research and theory suggest these correlations:

Related Trait	Correlation	Research Basis
Working Memory	r = 0.55	Context maintenance is memory-dependent
Comprehension	r = 0.38	Understanding structure aids reorientation
Persistence	r = 0.32	Persistent users try harder to resume
Patience	r = 0.28	Recovery takes time; patient users invest it
Reading Tendency	r = 0.25	Readers use text cues for recovery

Interaction Effects

Interrupt Recovery x Working Memory: Combined high values create maximally context-resilient users
Interrupt Recovery x Low Patience: Users may have recovery ability but not time patience to use it
Interrupt Recovery x Comprehension: High recovery + low comprehension = can find their place but may not understand current step

Persona Values

Persona	Interrupt Recovery Value	Rationale
power-user	0.75	Skilled at context-switching; uses environmental cues effectively
first-timer	0.35	Lacks schema for interpreting recovery cues
elderly-user	0.40	Working memory challenges impede context retention
impatient-user	0.45	May have ability but doesn't invest effort to recover
mobile-user	0.50	Moderate; mobile users frequently interrupted
screen-reader-user	0.55	Developed coping strategies for non-visual navigation
anxious-user	0.35	Anxiety impairs working memory and recovery
multi-tasker	0.70	Practiced at context-switching

UX Design Implications

For Low Interrupt Recovery Users (< 0.4)

Auto-save everything: Persist form data frequently and automatically
Session persistence: Don't timeout sessions aggressively
"Welcome back" states: Detect returning users and restore context
Prominent progress indicators: Make "where you are" unmissable
Breadcrumb navigation: Clear path back to current location
Unsaved changes warnings: Prevent accidental navigation away
Email/save progress links: Allow explicit progress saving

For High Interrupt Recovery Users (> 0.7)

Minimal recovery friction: Don't force re-authentication unnecessarily
Smart defaults: Pre-fill likely values based on previous session
Quick resume options: "Continue where you left off" buttons
Tab state preservation: Maintain state across browser sessions
History navigation: Support effective use of back button

Environmental Cue Best Practices

Cue Type	Implementation	Recovery Benefit
Progress indicators	Step X of Y, progress bars	20% faster recovery
Breadcrumbs	Clickable path hierarchy	15% faster recovery
Form persistence	Save partial form data	40-60% less re-entry
Descriptive titles	Page-specific, goal-oriented	10% faster orientation
Scroll restoration	Return to scroll position	Immediate context recovery
Visual state	Expand/collapse states preserved	Reduces re-navigation

Bibliography

Adamczyk, P. D., & Bailey, B. P. (2004). If not now, when? The effects of interruption at different moments within task execution. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 271-278. https://doi.org/10.1145/985692.985727

Altmann, E. M., & Trafton, J. G. (2002). Memory for goals: An activation-based model. Cognitive Science, 26(1), 39-83. https://doi.org/10.1207/s15516709cog2601_2

Czerwinski, M., Horvitz, E., & Wilhite, S. (2004). A diary study of task switching and interruptions. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 175-182. https://doi.org/10.1145/985692.985715

Iqbal, S. T., & Horvitz, E. (2007). Disruption and recovery of computing tasks: Field study, analysis, and directions. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 677-686. https://doi.org/10.1145/1240624.1240730

Mark, G., Gonzalez, V. M., & Harris, J. (2005). No task left behind? Examining the nature of fragmented work. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 321-330. https://doi.org/10.1145/1054972.1055017

Mark, G., Gudith, D., & Klocke, U. (2008). The cost of interrupted work: More speed and stress. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 107-110. https://doi.org/10.1145/1357054.1357072

Monk, C. A., Trafton, J. G., & Boehm-Davis, D. A. (2008). The effect of interruption duration and demand on resuming suspended goals. Journal of Experimental Psychology: Applied, 14(4), 299-313. https://doi.org/10.1037/a0014402

Trafton, J. G., Altmann, E. M., & Ratwani, R. M. (2011). A memory for goals model of sequence errors. Cognitive Systems Research, 12(2), 134-143. https://doi.org/10.1016/j.cogsys.2010.07.010

Copyright: (c) 2026 Alexa Eden.

License: MIT License

Contact: [email protected]

Trait InterruptRecovery