Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Aug 29;9(8):e106553.
doi: 10.1371/journal.pone.0106553. eCollection 2014.

Echoic Memory: Investigation of Its Temporal Resolution by Auditory Offset Cortical Responses

Free PMC article

Echoic Memory: Investigation of Its Temporal Resolution by Auditory Offset Cortical Responses

Makoto Nishihara et al. PLoS One. .
Free PMC article


Previous studies showed that the amplitude and latency of the auditory offset cortical response depended on the history of the sound, which implicated the involvement of echoic memory in shaping a response. When a brief sound was repeated, the latency of the offset response depended precisely on the frequency of the repeat, indicating that the brain recognized the timing of the offset by using information on the repeat frequency stored in memory. In the present study, we investigated the temporal resolution of sensory storage by measuring auditory offset responses with magnetoencephalography (MEG). The offset of a train of clicks for 1 s elicited a clear magnetic response at approximately 60 ms (Off-P50m). The latency of Off-P50m depended on the inter-stimulus interval (ISI) of the click train, which was the longest at 40 ms (25 Hz) and became shorter with shorter ISIs (2.5∼20 ms). The correlation coefficient r2 for the peak latency and ISI was as high as 0.99, which suggested that sensory storage for the stimulation frequency accurately determined the Off-P50m latency. Statistical analysis revealed that the latency of all pairs, except for that between 200 and 400 Hz, was significantly different, indicating the very high temporal resolution of sensory storage at approximately 5 ms.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.


Figure 1
Figure 1. Schematic illustration of auditory stimuli.
A click train was presented binaurally at 25 to 400 Hz (Experiment 1) or at a random frequency of 100 and 200 Hz (Experiment 2) for 1000 ms. Each click was generated as a single cycle of a 1000 Hz sine wave.
Figure 2
Figure 2. Effects of the click frequency on the latency of Off-P50m.
(A) Data from a representative subject. Equivalent current dipoles (ECDs) of the subject were superimposed on the subject’s own brain MR images. (B) Source strength as a function of time for the 200 Hz click train. (C) The off- and on-responses for 25 to 400 Hz click trains. The time course of each dipole is shown in the same color to that for the dipole location.
Figure 3
Figure 3. Grand-averaged source strength waveforms.
(A) Grand-averaged waveforms for each click train obtained from seven subjects. The off-response was enclosed by gray lines. (B) Note that the peak latency of Off-P50m was delayed in parallel with the stimulus frequency.
Figure 4
Figure 4. Relationship between the Off-P50m latency and stimulus frequency.
(A) The Off-P50m latency as a function of the ISI of the click train. The right column shows the latency difference of Off-P50m for each click sound relative to that for the 25-Hz train. Dotted lines indicate theoretical values. (B) Schematic illustration of the relationship between the latency of Off-P50m and ISI. Note that the delay in P50m latency reflects the click interval and the difference calculated by subtracting ISI from the P50m latency for each click is constant (right column).
Figure 5
Figure 5. Characteristics of the offset P50 component.
(A) Grand-averaged waveforms of Off-P50m in Experiment 2. Only an ambiguous Off-P50m was evoked by the 50/100 Hz random frequency click train. Note that the entire waveform for the random frequency click train (yellow) was later than that for the 50-Hz train. (B) Grand-averaged Off-P50m waveforms in Experiment 3 showed that the effects of attention on Off-P50m were negligible. (C) Grand-averaged waveforms of Off-P50m in Experiment 4. Note that the latency of Off-P50m was not altered by different sound pressure levels in both hemispheres (See also Fig. 3). (D) Auditory brain stem response (ABR) waveforms from one subject for click trains at 50 and 100 Hz. Upper waveforms were obtained with the start phase and end phase of the 100 Hz click train. Lower waveforms were obtained with the 50 Hz click train. Each waveform was an average of 1000 epochs and the waveforms of three sessions were superimposed. Right columns show simultaneously-recorded P50. Note the clear Off-P50 component in spite of the lack of any activity in ABRs elicited by the off event.

Similar articles

See all similar articles

Cited by 7 articles

See all "Cited by" articles


    1. Atkinson RC, Shiffrin RM (1968) Human memory: A proposed system and its control processes. The psychology of learning and motivation 2: 89–195.
    1. Winkler I, Cowan N (2005) From sensory to long-term memory: Evidence from auditory memory reactivation studies. Exp Psychol 52: 3–20. - PubMed
    1. Inui K, Urakawa T, Yamashiro K, Otsuru N, Takeshima Y, et al. (2010b) Echoic memory of a single pure tone indexed by change-related brain activity. BMC Neurosci 11: 135. - PMC - PubMed
    1. Cowan N (1984) On short and long auditory stores. Psychol Bull 96: 341–370. - PubMed
    1. Lu ZL, Williamson SJ, Kaufman L (1992) Behavioral lifetime of human auditory sensory memory predicted by physiological measures. Science 258: 1668–1670. - PubMed

Grant support

The authors have no support or funding to report.