Human startle reflex: technique and criteria for abnormal response

S. Chokroverty, T. Walczak, W. Hening

Research output: Contribution to journalArticlepeer-review

65 Scopus citations


Because quantitative norms for the normal audiogenic startle response to repeated stimuli have not been previously reported, we now describe a technique for eliciting the startle response and anlaysing its habituation with repeated stimuli. We used binaural 105 dB tones delivered in 5 blocks of 4 tones. Successive blocks were separated by a 5 min period without tones stimuli and had progressively shorter inter-stimulus intervals (ISIs) beginning with 5 min in the first block and reducing to 1 min in the final, fifth block. We contrast the response and its habituation in a group of 8 normal subjects with that in a patient with clinically exaggerated startle. Based on the differences observed, we propose that the following criteria may be used to ascertain an abnormally increased startle response: (1) excessive duration of the myogenic response; (2) persistence of extracranial responses after the initial two blocks of stimuli; and (3) reduced habituation of the response (as measured by decreases in response duration and in the area under the curve of rectified EMG for the orbicularis oculi myogenic response). Our patient was abnormal on each of these measures. This result is consistent with past qualitative reports which have indicated that abnormal startle is associated both with excessive startle and with subnormal habituation. Study of further patients with hyperekplexia will be necessary to either confirm our data or modify our proposed criteria.

Original languageEnglish (US)
Pages (from-to)236-242
Number of pages7
JournalElectroencephalography and Clinical Neurophysiology/ Evoked Potentials
Issue number4
StatePublished - Aug 1992


  • Habituation
  • Hyperekplexia
  • Startle reflex

Fingerprint Dive into the research topics of 'Human startle reflex: technique and criteria for abnormal response'. Together they form a unique fingerprint.

Cite this