Are you an evening or a morning person?

The journal Science has published an article entitled Homeostatic Sleep Pressure and Responses to Sustained Attention in the Suprachiasmatic Area, whose main author is Christina Schmidt, FNRS candidate at the ULg’s Cyclotron Research Centre. For the first time, functional magnetic resonance imaging has been used to study the influence of the chronotype on the way the brain functions in individuals faced with a cognitive task. The results: appreciable differences in terms of the level of activity in certain areas of the brain between ‘morning types’ and ‘evening types.’

All living beings possess an internal biological clock which allows them to adapt to periodical variations in their environment, in particular the changeover from day to night time. In humans, as in mammals, it comes in the form of an aggregate of neuronal cells located in the antero median section of the hypothalamus. Its name? The suprachiasmatic nucleus.

Its metabolic and electro-physiological activity has a natural rhythm over a period slightly greater than 24 hours. At least for the majority of us. This is what are called circadian rhythms: circa: around, diem: day). Nonetheless, in normal conditions elements of the physical and social environment named ‘time synchronisers’ recalibrate our biological clock to a scale of 24 hours. The principal one is constituted of variations in brightness caused by the changeover from day to night. But there are others, such as work schedules and meal times. Since its beginnings, chronobiology has studied numerous physiological variables from the perspective of circadian rhythms: body temperature, the production of cortisol, alertness, heart rate, mood, etc. It is also interested in cognitive performances, whose variable nature in terms of the time of the day has been well demonstrated.

A more subtle analysis nevertheless indicates that many factors need to be taken into consideration: first of all, the circadian rhythms; next, sleep pressure, or the need to sleep, linked to the number of hours the subject has remained awake. These two processes interact in concert or against each other according to the moment of the day to ensure an adequate awake-sleeping rhythm. The different inter-individual differences which modulate this interaction between circadian rhythm and sleep pressure also need to be taken into account. in fact there exist specific ‘chronotypes’ which can be inscribed on a Gaussian curve: 66% of the population is said to be neutral, in other words neither ‘morning’ or ‘evening’ types, whilst the remaining 34% are divided harmoniously into categories which we can term ‘moderate morning’, ‘moderate evening, ‘extreme morning’ (5%) and ‘extreme evening’ (5%) types.

Two processes

Most research work related to variations in cognitive performance over the course of the day has focused on the influence of the homeostatic process, expressing the need to sleep, and the circadian process which, in the words of Christina Schmidt, FNRS candidate at the University of Liège, constitutes a clock of the propensity to sleep and modulates the thresholds of wakefulness over the course of the day.

The ‘game’ the two processes become engaged in is not an innocent one, because it determines the daily variations of fatigability or overall attentiveness, and by the same token can affect performance levels over a series of cognitive tasks: arithmetical, reaction times, short term memory, etc. (Read the multiple memory). Studies carried out up until now have enabled a separating out of the influence of these two components, homeostatic and circadian, but have usually only looked at tasks involving relatively simple cognitive processes, such as speed and vigilance tests.

In these conditions the performances attained by the subjects coincide with the circadian body temperature curve, which constitutes a faithful reflection of the state of wakefulness over the course of the nycthemeron. In the absence of well established data we nevertheless do not know if more evolved tasks, relying for example on memory or executive functions, are housed at the same lodgings.

Several studies have looked into the links between the chronotype and cognitive performance. They reveal that the latter is at its best or, conversely, at its worst, at a precise moment during the day, itself a function of the subject’s neutral, early rising or evening profile. As Christina Schmidt indicates, this chronotypical susceptibility has been observed for tasks exploring various aspects of our executive functions, but seems to affect the production of routine or conditioned responses less, or even not at all.

Swiss experiments

Beyond the homeostatic process, circadian rhythm and the chronotype, a fourth factor needs to be taken into account: age. De facto, over the course of ageing, human beings tend to become more and more ‘morning people.’ ‘The interaction between cognitive variables and the moment of the day changes over the course of ageing, young and old subjects differing significantly in their optimal circadian peak,’ states Christina Schmidt.

Very interesting experiments have been carried out at the University of Basel’s Centre for Chronobiology, led by Professor Christian Cajochen, a laboratory in which Christna Schmidt carried out a one and a half year training programme (2005-06). Her work was carried out in a very controlled environment, and one which was very restrictive for the volunteers who took part in them. During a full week each of them had to stay in a room in which the eating and lighting conditions, as well as the body posture to be assumed (half lying down), were rigorously defined.

The volunteers were offered a so called procedural task, in this case a motor skills test. They were asked to place themselves in front of a computer screen on which had been fixed dots below which a visual clue could appear. A button corresponded to each of the dots. The task constituted in pushing the right one as quickly as possible when the visual prompt appeared on the screen.

Every volunteer was submitted to two experimental procedures. In the first they were deprived of sleep for 40 consecutive hours; as for the second, it hinged on half an hour siestas every 90 minutes. For the first experiment the goal was to increase sleep pressure beyond a threshold which allowed for the individual impact of circadian and homeostatic mechanisms on cognitive performance to be singled out.

The two procedures interact on a permanent basis, let us remind ourselves. For example, at 19.00, sleep pressure is already heightened and in itself could push people to go to sleep. But the wake promoting circadian signal encourages people to stay awake, in a way counteracting the activity of the homeostatic mechanisms. Also, the cognitive performance tests carried out during a normal day did not allow the exact influence of the two components studied to be singled out. Nevertheless, after being awake for 40 hours, sleep pressure reaches a form of saturation. In these conditions, if it was observed that the performance of the subjects improved at certain moments of the day despite an ever increasing need to sleep, here the imprint of a circadian peak can be seen.

That is what the experiment carried out in Basel emphasized in an unequivocal manner: there do indeed exist circadian variations in performing the procedural task demanded. The researchers did not observe a linear decline in the performances according to sleep pressure; on the contrary modulations connected to variations in circadian wakefulness, including body temperature, were observed. In other words the performances were better during the ‘right circadian phase’, no matter the sleep pressure at work. Similar results were obtained in the second experimental procedure, where the sleep pressure was monitored in a different manner, through a series of siestas prescribed to the experiments’ participants.

Towards 6 o’ clock in the morning, body temperature is minimal and the secretion of melatonin maximal. In laboratory controlled conditions it is at this moment that performances concerning simple cognitive tasks are the worst for subjects whose chronotype could be called neutral. They then improve over the course of the day in parallel with the rise of body temperature, whose maximum measurement is reached around 22.00. It remains to be discovered what exactly happens in daily life, in which several factors play a part: the levels of the day’s light, the motivations, the journeys and the gestures carried out by the subject, etc.

The paths of extremes

For more complex tasks the data remains vague, as the results of the experiments revealed a lot of variability. Should one call into question the methodological reasoning? Maybe. Even so, an experiment carried out by Christian Cajochen and his University of Basel team seems to have underlined a pattern of performances analogous to that which had been established within the tests looking into the aforementioned procedural task. The second experiment followed the broad lines of the first one – a strictly controlled environment, sleep deprivation, etc. - but only looked at elderly people. The latter were asked to direct themselves through ‘paper-pencil’ versions of labyrinths. It appeared that the harder the task was, the more marked were the performances’ sensitivity to circadian influence. The work carried out at the University of Basel called on so-called ‘neutral’ subjects in terms of chronotype. Otherwise, in the perspective that they had fixed themselves, it would have been almost impossible to disentangle the web.

At the University of Liège, Christina Schmidt took a different path. In effect, the work which enabled her to figure as the major author of the article entitled Homeostatic Sleep Pressure and Responses to Sustained Attention in the Suprachiasmatic Area(1), published on the 24th of April 2009 in the journal Science, put on the stage more ‘marginal’ subjects, in other words people who are ‘extreme morning types’ or ‘extreme evening types.’ In addition they were young individuals (18-30 years old). Why? Because beyond the age of 30 the parameters of sleep tend to change – with age an ever increasing propensity to ‘become morning types’ manifests itself. Two groups of 16 volunteers were established, one made up of ‘extreme morning types’, the other of ‘extreme evening types.’ ‘The selection was carried out on the basis of a questionnaire to which 6000 people responded,’ reports Christina Schmidt. ‘Different variables were taken into consideration, because we need to be careful of misleading appearances. Thus, certain people get up early because they are obliged to due to their work schedule. The people we were interested in for the ‘extreme morning’ were not these people, but those who rise early spontaneously, without an alarm clock, even at the weekend. In the same way, the ‘extreme evening types’ are people who, when they have the opportunity, watch television until three or four o’ clock in the morning and do not leave their beds before around 11 o’ clock in the morning.’

Another special feature of the work reported in the 24th of April issue of Science is that it calls on functional magnetic resonance imaging (fMRI), whereas all the studies previously carried out on the relationships between biological rhythms and cognition were based on behavioural issues.

The influence of the chronotype

At the heart of a collaboration between the University of Liège (Cyclotron Research Centre, department of Cognitive Sciences) and the University of Basel (Centre for Chronobiology), the programme managed by Christina Schmidt under the supervision of Professor Philippe Peigneux (ULB) and Fabienne Collette (ULg), with the help of Pierre Maquet (ULg) et de Christian Cajochen, has thus delivered its first results. They relate to a visual attention task bringing into play the subject’s reaction times. What did it consist of? The volunteers had to focus on a cross on a computer screen. From time to time, and in random fashion, the cross disappeared and gave way to a digital countdown. It was thus a question of pushing a button as quickly as possible in order to stop it. At the same time the subject’s brain activity was recorded by fMRI. The test, which was carried out twice over the course of a day, lasted 10 minutes and numbered 90 events (appearances of the countdown). They did not occur at just any old time. No, the challenges were offered to each volunteer one and a half and ten and a half hours after the time of day they usually wake up at.

This original procedure offers two essential advantages. On the one hand it enables an equalling out between the subjects of the number of hours spent in a state of wakefulness and the sleep inertia which follows waking up, two variables which can influence the results. On the other hand, it takes into account the fact that the evening and morning subjects could be differentiated by their state of vigilance at specific times of the day. ‘Without controlling this aspect, we could not exclude the hypothesis that the chronotypical variability of the performances concerning the prescribed task would only be secondary to a variation in the levels of vigilance,’ explains Christina Schmidt.

(1) Schmidt, C., Collette, F., Leclercq, Y., Sterpenich, V., Vandewalle, G., Berthomier, P., Berthomier, C., Philipps, C., Tinguely, G., Darsaud, A., Gais, S., Schabus, M., Desseilles, M., DangVu, T., Salmon, E., Balteau, E., Degueldre, C., Luxen, A., Maquet, P., Cajochen, C., & Peigneux, P. (2009). Homeostatic Sleep Pressure and Responses to Sustained Attention in the Suprachiasmatic Area. Science 324, 516.

The results of the experiment? There was no observed difference in the levels of performance or cerebral activity between the two categories of subject (‘extreme morning’ and ‘extreme evening’ types) when the test was carried out 1h30 after waking, in other words at a moment when sleep pressure is low. On the other hand, 10h30 after waking, under greater sleep pressure, the attention giving performance improves amongst subjects who have an ‘extreme evening’ chronotype, which is not the same case for their ‘extreme morning’ counterparts. At the same time the suprachiasmatic nucleus (SCA – circadian biological clock) and the locus coeruleus (LC), two regions of the brain anatomically inter-connected and heavily involved in the circadian signal which promotes wakefulness and regulates our vigilance during the wakeful state, experienced an increase in activity amongst the evening subjects.

‘On the basis of this data, our hypothesis was that sleep pressure accumulates in a quicker manner for ‘the morning subjects’, in such a way that they are more tired than ‘the evening subjects’ when they have spent the same number of hours in a waking state,’ points out Christina Schmidt. There exists an excellent indicator of sleep pressure: the density of slow waves (recorded by electroencephalography) during the first sleep cycle. And, as a result, it appeared that the subjects equipped with a morning chronotype where characterised by a strong density of such waves at the beginning of the night.

From which the researchers’ conclusion: ‘Sleep pressure is inversely linked to the level of activity in the region of the suprachiasmatic nucleus during the vigilance task, showing for the very first time in humans that the activity of brain circuits responsible for circadian regulation is modulated by the homeostatic processes of sleep. This suggests that ‘morning subjects’ suffer more strongly than ‘those of the evening’ from the sleep pressure accumulated over the course of the day, a pressure which prevents the optimal expression of the wake promoting signal in the regions of the suprachiasmatic nucleus and the locus coeruleus.’

More complex tasks

The work of the Belgo-Swiss team have not stopped when presented with such a good pathway. Beyond the (simple) visual attention task, the researchers offered the two groups of 16 selected volunteers a task involving short term memory and another called conflict management, bringing into play the executive functions.

For the short term memory task the participants were placed in three different situations. In the first they had to indicate, using two buttons, one for ‘yes’ and one for ‘no’, if a letter that was presented to them on a screen corresponded to a predefined letter (T, for example). This was a control condition allowing the brain areas which were activated on seeing the letters to be determined, without a short term memory task being prescribed. In the second situation, where letters were made to appear successively, the participants were asked to decide if the letter they were seeing corresponded or not to the letter which was fixed on the screen two ‘times’ beforehand. Thus a mnestic load was added to the control condition. A load which was increased in the third situation, in which it was necessary to refer to the letter which appeared three ‘times’ beforehand. Here as well the tests were carried out, for each participant, 1h30 and 10h30 after their usual waking up time. The results (2), which have not yet been analysed in detail nor published, seem to corroborate the results received for the preceding test (visual attention). Nonetheless the differences between ‘morning subjects’ and ‘evening subjects’ no longer focuses on the reaction times but on the number of correct responses given.

(2) The results of the fMRI are in particular are in the process if being analysed.

As for the conflict management test, which was based on the use of the Stroop test (it was a question of inhibiting an automatic response in favour of a more controlled procedure), it delivered a strange truth: the performances of the ‘extreme morning types’ are equivalent to those of the ‘extreme evening types’, but the activity of certain areas of the brain involved in conflict management (insula, anterior cingulate cortex) increased with the waking time of the latter, and uniquely for them. Why? The researchers will make every effort to pierce this mystery. ‘If we had carried out the test 16 hours after waking up, for example, it is possible that the performances of the ‘evening subjects’ would have turned out to be superior to those of the ‘morning subjects,’ stresses Christina Schmidt. ‘From this we can imagine that the differences of activation within the brain would be reflected in the level of performances, as if after going up a level.’ As far as the suprachiasmatic nucleus and the locus coeruleus are concerned, the results confirm the data of the visual attention test, but with a lower statistical threshold.

Neuro-psychological evaluations

The work initiated at the University of Liège do not spring solely from the sphere of basic research, but also have the ambition to influence clinical practices. In effect, the neuro-psychological evaluations today carried out to explore the cognitive functioning of brain damaged patients generally tend not to take into account the existing links between the time of day the tests are carried out and the individual chronotype of the subject being examined.

Let us say that we want to follow up a patient over the long term in evaluating his cognitive performances every three months. If the tests take place sometimes in the beginning of the morning, sometimes at midday, sometimes at the end of the afternoon, for example, the measurements of the case’s development could be distorted. The same could be the case if, putting to one side the time of day, we submitted a person to a single test in order to analyse his results in the light of those of a control group. In addition, young and old subjects are placed on the same footing, whereas we know that the more we age, the more we swing towards a morning chronotype. ‘In certain tests, it has been shown that the differences in performance between these two types of people were erased when the young individuals were tested at a non-optimal moment and the elderly individuals were tested at the moment of their circadian peak,’ stresses Christina Schmidt on the other hand.