中文 |

Research Progress

Patients with Schizophrenia Do Not Show Impaired Temporal Attention

May 07, 2015

How many times you remember celebrating your experiment has showed the same data as previous studies during your graduate study, and how upset you were when you didn’t get the results that everyone was expecting?

One of the most important aspects of any branch of sciences is reproducibility, which means any finding by one study should be reliably replicated by others. However, what happened if after checking everything in your experiment and you still find that your data is different? Recently, there have been several famous incidences where findings that were thought to be important and groundbreaking got published in very high profile journals, but later failed to be replicated by independent studies. Many of these incidences are not due to data fabrication, which nowadays the public is well aware of, but less obvious flaws in their methodologies. Can we learn something when data does not reproduce?

Dr. SU Li from Department of Psychiatry at University of Cambridge in collaboration with Prof. CHAN Raymond from Institute of Psychology of Chinese Academy of Sciences have questioned a series of studies published over the last 13 years in many highly influential journals, such as Schizophrenia Bulletin (top 5% of all psychiatry journals) over an important issue regarding the temporal attention deficit in schizophrenia.

The previous studies in the literature have supported the view that schizophrenia patients have impaired temporal attention beyond and above their existing perception impairments. These studies all used a classic experimental paradigm called the Attentional Blink, which is designed to probe temporal attention. In those studies, a mathematical method was used in the post hoc analysis trying to remove the influence of the basic perception difference between schizophrenia patients and their controls. This method is called suppression ratio. By using this mathematical trick, these studies have claimed that the suppression ratio, which was supposed to remove the visual perception variations, should reflect the pure deficit in attention. As such, they all reported a marked decrease in attention in patients with schizophrenia.

Researchers revisited the issue, but this time used a more rigorous experimental design to control for the visual perception differences between the schizophrenia patients and the controls, namely matching the perceptual difficulty of the task. In this way, they did not rely on the use of suppression ratio, and found no evidence for an impaired temporal attention. Instead, an increased level of temporal binding error was associated with the schizophrenia patients. Using computational modeling approach and more careful investigation of the mathematical method used in the previous studies, researchers have concluded that the suppression ratio used in almost all previous studies has systematically exaggerated the degree of attentional impairment in schizophrenia. This flaw in methodology explains why the original findings were not replicated.

Failed to replicate the previous findings, the study however showed an unreported but expected impairments in schizophrenia, that is, the patients with schizophrenia tend to have poorer temporal binding accuracy. This is illustrated by the fact that when two visual stimuli were presented within a very short time interval (around 100ms), schizophrenia patients have greater difficulty to tell in which order the stimuli were presented. This finding is consistent with the previous literature on temporal perception deficit in schizophrenia. The authors argue that inability to encode sensory information in the precise order in which they are occurring in the environment can lead to serious problem in both perception and reasoning. To some extent, such a deficit contributes to many symptoms in schizophrenia such as hallucination.

The study demonstrated clearly how scientists could uncover important message when their data are different from existing literature. It also reminds us the danger of misusing post hoc data analysis methods, in particular in situations where the error “improves” the results. This study also encourages students and young scientists to challenge existing theories and data by applying more rigorous methodology as well as mathematically modeling.

This work entitled “Temporal perception deficits in schizophrenia: integration is the problem, not deployment of attentions” was published online in Scientific Reports.

This study was supported by the open competitive grant of the key Laboratory of Mental Health, Institute of Psychology, NIHR Biomedical Research Centre and Biomedical Research Unit, NHS Foundation Trust, the National Science Fund China Young Scholar Investigator Award and the Knowledge Innovation Project of the Chinese Academy of Sciences.

Contact Us
  • 86-10-68597521 (day)

    86-10-68597289 (night)

  • 86-10-68511095 (day)

    86-10-68512458 (night)

  • cas_en@cas.cn

  • 52 Sanlihe Rd., Xicheng District,

    Beijing, China (100864)

Copyright © 2002 - Chinese Academy of Sciences