In this study, developmental changes in duration of the icon (visual sensory store) were investigated with three converging tachistoscopic tasks. (1) Stimulus interuption detection (SID), a variation of the two-flash threshold method, was performed by 29 first- and 32 fifth-graders, and 32 undergraduates. Icon duration was estimated by stimulus interruption that resulted in 75% performance. Results suggest that the icon lasts longer in younger subjects. (2) Letter detection (LD), an adaptation of the Estes and Taylor detection method, estimated effective icon duration as the shortest interval between the stimulus field and noise mask that resulted in equivalent-to-unmasked performance. Data were collected over seven sessions with ten 6- to 11-year-old females. No strong relation between age and iconic duration was apparent. (3) Judgement of cyclic stimulus disappearance, fashioned after Haber's direct method consisted of a letter matrix, alternating with a blank field. As the blank field increased in duration, the LD subjects judged when the letters seemed to disappear and when the letters seemed to remain on continuously as the blank field duration was decreased. The results indicate no strong evidence for a change in icon duration with age. Methodological issues were examined in comparing the results of the three approaches. The age difference in SID can be attributed to factors other than icon duration. The evidence favors the view that icon duration is constant across age. (Author/MS)