A Study of Decision-Making Across Time

Participants are randomly assigned to one of two linguistic framing conditions adapted from prior research on intertemporal decision-making. The conditions differ only in how otherwise equivalent reward information is linguistically presented. The precise wording of the conditions is withheld until completion of the study. Monetary amounts, delay periods, task structure, and financial incentives are held constant across conditions.

2026-10-01

2027-01-30

The primary outcome is the participant’s choice of the delayed reward rather than 100 tokens today in each intertemporal-choice round. A key participant-level summary outcome is the switch point for each delay horizon, coded on a 10-category ordered scale: Line 1 = always choose the delayed option; Line 2 = switch at 105 tokens; Line 3 = 110; Line 4 = 115; Line 5 = 120; Line 6 = 125; Line 7 = 130; Line 8 = 135; Line 9 = 140; and Line 10 = always choose the immediate option. A lower switch-point code indicates greater patience.

For each round, the delayed-choice outcome equals 1 when the participant chooses the larger delayed reward and 0 when the participant chooses 100 tokens today. The descriptive proportion choosing the delayed reward is calculated separately for each randomized condition from all recorded choices of eligible participants, without excluding multiple-switch participants. Within each delay horizon, delayed amounts are presented in ascending order. A participant is classified as a multiple-switch participant if their choices change between immediate and delayed options more than once within any delay horizon. All other analyses exclude any such participant in full: none of their decisions from any horizon or other outcome observations are included. For retained participants, the switch point is coded on a 10-category ordered scale: 1 for always choosing the delayed option; 2, 3, 4, 5, 6, 7, 8, or 9 for switching to the delayed option at 105, 110, 115, 120, 125, 130, 135, or 140 tokens, respectively; and 10 for always choosing the immediate option. Always-delayed and always-immediate patterns are not classified as multiple-switch patterns.

Decision response time; perceived psychological distance; and the participant’s written description of the reasoning used for the final decision.

Response time is measured in milliseconds from the moment the choice options become available until a choice is recorded. Psychological distance is measured using three seperate scenarios (1 week, 4 week, 12week).

This study is a two-arm, individually randomized online conceptual replication and extension of prior research on linguistic framing and intertemporal decision-making. The planned sample is 400 participants recruited through Prolific. Allocation to the two conditions is determined by individual randomization with equal probability, so the realized numbers in each condition may differ. The study examines whether previously documented effects of linguistic framing generalize to a different language, participant population, and online implementation. After providing consent and completing an audio check, participants receive instructions and complete 24 incentivized choices between rewards available at different times. Participants are randomly assigned to one of two linguistic framing conditions, while the underlying economic content and incentives are held constant. One decision is randomly selected for bonus payment. The study records participants’ choices and decision times, followed by measures concerning their perceptions, reasoning, and demographic characteristics.

Not available

Computerized random assignment implemented in Qualtrics. The survey flow assigns each participant with equal probability to one of the two linguistic framing conditions. Qualtrics also randomizes the order of the three delay-horizon blocks.

Individual participant.

No

400 individual participants (the randomization is not clustered).

400 participants, each completing 24 decision rounds (up to 9,600 participant-round observations).

The planned sample is 400 participants, with 1:1 assignment probabilities for linguistic framing conditions A and B (a target of 200 per condition). The realized number in each condition is determined by individual randomization and may differ from 200.