Chemistry 151 Week 4 – Critically Analyzing Scientific Data
College of the Canyons Fall 2020
Name ______________________
Date ________________ Section_________________________________
Critically Analyzing Scientific Data
Activity adapted from Miller, D. M; Chengelis Czegan, D.A.Integrating the Liberal Arts and Chemistry: A
Series of General Chemistry Assignments to Develop Science Literacy. J. Chem. Ed, 2016, 864-869
In today’s fast-paced, technology-driven, 24-hours news cycle society, it seems information is coming at
you faster than you can keep up. Much of this information is “scientific” news, stories, or advances and can
sound very legitimate…but that is not always the case. As an informed listener or reader, you need to be
able to take in the information being presented, critically and objectively analyze it, and make your own
judgement as to the validity of the conclusions that are being reported.
In science there are many factors that affect whether a news source, website, or story is credible. This
activity is designed to help you assess what information should be taken seriously versus information being
hyped to help attain some other purpose. You will apply those skills to analyze some real-life, current
scenarios.
Hallmarks of Reliable Scientific Research
When scientists observe some phenomenon that is interesting to them they often use the scientific method
to learn more about that process. The scientific process consists of the following steps:
Observation
This process provides a framework to analyze new findings, but it does not ensure the validity of any
findings. A valid scientific study is characterized by additional features that serve to challenge and
scrutinize any new findings prior to publication. These characteristics are:

  1. Representative samples (large n, reproducible data)
  2. Reproducibility (control groups, cause vs correlation, biases & placebo effects)
    It is unavoidable that there will be some amount of variation in results from a scientific experiment. This
    arises from random error sources, which are outside of the control of the researcher. This could be changes
    in the environment (temperature fluctuations for example) or just the inability to exactly repeat precisely
    how you took a measurement. Random error affects scientific data’s reproducibility or precision: the
    ability to obtain the same result every time an experiment is performed. The good news about random error
    is that it usually averages out statistically (for every temperature fluctuation where it got a little warmer,
    there will likely be a fluctuation where it got a little cooler).
    If too few measurements are used to draw conclusions, random errors present in the experiments may be
    misinterpreted as true phenomena, and a scientist may draw unwarranted conclusions from the data. This
    means when you are hearing about a new scientific discovery, information, or finding, if it is unclear how
    many times the experiment was repeated, or even worse if it was clearly not repeated at all, or very small
    sample sizes were used, then you should be skeptical about the results.
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
  3. Peer Review (valid interpretation of results, well-designed and performed studies)
    This process of peer-review helps to ensure that all published data is valid, significant, and original.2 This
    means that the science that comes out of this process is more than just some scientist’s opinion; it’s been
    analyzed, scrutinized, and supported by a variety of scientists in the field.
    When a group of researchers believe that they have data and information worth sharing with the world, they
    will submit their paper to be published in a journal. These are different than magazines, newspapers or
    online articles because while those articles are hopefully thoroughly researched, they are not critiqued by
    other, non-affiliated, objective experts in the same field. Not all journals are peer- reviewed either, but those
    that are send the papers they receive to a team of experts to read the paper before they agree to publish it.
    These experts critically examine the scientific data presented to determine if it is:
    • valid/credible
    • significant
    • original2
    If enough of the peer-reviewers agree that the science presented is credible, reproducible, and valid, then
    the journal will likely agree to publish the results. If the peer-reviewers find issues, concerns, or other errors
    in the data, the journal will not publish the study as presented. Often, the journal requests that the scientists
    edit or amend the original research – often requiring that additional experiments be carried out! – before
    agreeing to publish the paper. In a sense, peer-reviewers are the referees for scientific research and
    conclusions, and if you find information in a peer-reviewed journal, you can have more confidence in the
    validity and accuracy of the data and conclusions that are being presented.
    One limitation to the peer-review process may come if someone is trying to publish truly brand-new, neverbeen-seen-before data. If there are no other obvious ‘experts’ in a field, it may be hard to find reviewers
    who can critically and objectively assess the data and the conclusions2. Usually as long as all of the data is
    clearly presented, and logically and honestly analyzed and discussed, even non-expert reviewers will be
    able to recognize it as valid and credible science and support its publication.
    For more information on peer review, see the handout listed in the references (2).
    Procedure
    Part I: Recognizing Reliable vs Unreliable Evidence
    Begin by reading the attached handout titled, “A Rough Guide to Spotting Bad Science”1 and answer the
    following questions:
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
  4. From the twelve points outlined on the handout, rank the three points that you feel are most detrimental
    to the validity of a scientific study. Then discuss your choices with your partner. Whenyou and your
    partner agree on the top three points, write them in the table below and write 1-2 sentences explaining
    why you ranked each as very detrimental.
  5. The handout mentions that both a control group and a blind study are ways to improve the reliability of
    a scientific study. What is the difference between a control group and a blind study?
    Point Explanation
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
  6. The difference between correlation and causation was discussed as well. In your own words, what is
    the difference between the two, and how do you think you can try to assess whether something you’re
    reading is correlation or causation?
  7. What is peer review? What is peer review important?
  8. What is one challenge with peer review?
    Part II: Searching and Assessing Reliable Sources
    Visit the following websites and based on what you’ve learned about identifying credible, reproducible,
    and peer-reviewed data, rank them on a scale from 1-5 for their reliability (1 is the least reliable source, 5
    the most). All of the sources deal with the safety of the artificial sweetener, Aspartame.
    x https://www.fda.gov/food/ingredientspackaginglabeling/foodadditivesingredients/ucm397725.ht m
    x http://articles.mercola.com/sites/articles/archive/2011/11/06/aspartame-most-dangerous- substance-addedto-food.aspx
    x https://en.wikipedia.org/wiki/Aspartame
    x http://www.equal.com/products/equal-original/
    x http://thepopularman.com/aspartame-detox-and-withdrawal/
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
  9. Use the table provided below to rank the websites, and provide justification for your ranking. Include
    in the justification any of the 12-points to look for that may be relevant, as well any other factor that
    you used in your decision.
    Website Ranking (1-5) Justification
    FDA
    Mercola
    Wikipedia
    Equal
    Popularman
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
    Part III: Journal Article Critique
    Read the attached article, published in Physiology and Behavior, and answer the questions that follow:
  10. Article analysis
    a. What was the goal of the research?
    b. What conclusions did the researchers draw from their study?
    c. What evidence did the scientists present that supported their claims?
  11. Article Critique
    a. Was this article published in a peer-reviewed journal?
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
    b. Identify any of the 12-points/criteria used to assess scientific findings that you see arise in this article.
    Provide the specific example from the article that matches the particular point, and in 1-2 sentences explain
    whether this was something that was positive (they avoided one of the 12 common faults) or negative (they
    utilized one of the 12 common faults in their findings).
    c. Read the ‘Acknowledgements’ carefully – who funded this research? What do youthink that means about
    the objectivity/validity of the study?
    d. Do a quick internet search to try and find who founded/funded the EuropeanHydration Institute. Does this
    make you more or less confident in the scientific findings?
    Chemistry 151 Week 4 – Critically Analyzing Scientific Data
    College of the Canyons
    e. Based on your response to questions (a) – (e), make a conclusion about the overall validity of this scientific
    data and the author’s conclusions. Make sure you fully explain your conclusion and how you came to it.
  12. Do you think that just because a scientific study was funded or initiated by a company/corporation with a
    potential conflict of interest, that it should immediately be assumed to invalid and skewed? If yes, explain
    why you feel this way. If not, explain what you think it does mean to you, the reader, if you discover a
    scientific study has ties toanother organization.
    Mild hypohydration increases the frequency of driver errors during a
    prolonged, monotonous driving task
    Phillip Watson a,c,
    ⁎, Andrew Whale a,b
    , Stephen A. Mears a
    , Louise A. Reyner a,b
    , Ronald J. Maughan a
    a School of Sport, Exercise and Health Sciences, Loughborough University, Leicestershire LE11 3TU, UK
    b Sleep Research Centre, Loughborough University, Leicestershire LE11 3TU, UK
    c Department of Human Physiology, Vrije Universiteit Brussel, Brussels B-1050, Belgium
    HIGHLIGHTS
    • Mild hypohydration has been shown to cause impaired cognitive function and altered mood.
    • This study reports an increase in driver errors with mild dehydration.
    • Error incidence increased over time, but occurred at a greater rate following fluid restriction
    • Higher subjective feelings of thirst, as well as impaired concentration and alertness were also apparent
    • Driver education programmes should also encourage appropriate hydration practices.
    article info abstract
    Article history:
    Received 13 November 2014
    Received in revised form 13 March 2015
    Accepted 13 April 2015
    Available online 16 April 2015
    Keywords:
    Cognitive function
    Dehydration
    Fluid balance
    Road traffic accident
    The aim of the present study was to examine the effect of mild hypohydration on performance during a
    prolonged, monotonous driving task.
    Methods: Eleven healthy males (age 22 ± 4 y) were instructed to consume a volume of fluid in line with
    published guidelines (HYD trial) or 25% of this intake (FR trial) in a crossover manner. Participants came to the
    laboratory the following morning after an overnight fast. One hour following a standard breakfast, a 120 min
    driving simulation task began. Driver errors, including instances of lane drifting or late breaking, EEG and heart
    rate were recorded throughout the driving task.
    Results: Pre-trial body mass (P = 0.692), urine osmolality (P = 0.838) and serum osmolality (P = 0.574) were the
    same on both trials. FR resulted in a 1.1 ± 0.7% reduction in body mass, compared to −0.1 ± 0.6% in the HYD trial
    (P = 0.002). Urine and serum osmolality were both increased following FR (P b 0.05). There was a progressive
    increase in the total number of driver errors observed during both the HYD and FR trials, but significantly
    more incidents were recorded throughout the FR trial (HYD 47 ± 44, FR 101 ± 84; ES = 0.81; P = 0.006).
    Conclusions: The results of the present study suggest that mild hypohydration, produced a significant increase in
    minor driving errors during a prolonged, monotonous drive, compared to that observed while performing the
    same task in a hydrated condition. The magnitude of decrement reported, was similar to that observed following
    the ingestion of an alcoholic beverage resulting in a blood alcohol content of approximately 0.08% (the current UK
    legal driving limit), or while sleep deprived.
    © 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
    (http://creativecommons.org/licenses/by-nc-nd/4.0/).
  13. Introduction
    Under ‘normal’ conditions, an individual's total body water (TBW)
    fluctuates throughout the day, but overall daily water balance is generally maintained through a series of interrelated factors which control
    intake and output of water. The homeostatic regulation of salt and
    water balance normally acts to limit excursions in TBW to no more
    than about 1% per day [24]. Nevertheless, there are several routinely encountered situations that act to either increase fluid losses (e.g. illness,
    exposure to heat/humidity, diuretics), or serve to restrict fluid
    intake (e.g. access to beverages and/or latrines). Over time, one, or a
    combination, of these factors results in the progressive reduction in
    TBW. The ensuing hypohydration causes a reduction in the circulating blood volume and an increase in plasma osmolality, which are
    typically proportional to the magnitude of decrease in TBW [32].
    Populations at particular risk of hypohydration are the very young,
    those engaged in professions where fluid homeostasis is regularly
    Physiology & Behavior 147 (2015) 313–318
    ⁎ Corresponding author. Fax: +32 26292876.
    E-mail address: pwatson@vub.ac.be (P. Watson).
    http://dx.doi.org/10.1016/j.physbeh.2015.04.028
    0031-9384/© 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
    Contents lists available at ScienceDirect
    Physiology & Behavior
    journal homepage: www.elsevier.com/locate/phb
    challenged and the elderly. Limited data are available on the prevalence of hypohydration, but there is evidence to suggest that this
    may be relatively common among sections of the elderly population
    [24].
    Mild hypohydration can cause symptoms such as headache, weakness, dizziness and fatigue, and generally makes people feel tired and
    lethargic, with lower self-reported ratings of alertness and ability to
    concentrate [36]. Body water losses have been shown to impair performance in a variety of tests of both physical and mental performance. Evidence suggests that either starting exercise in a hypohydrated state, or
    allowing hypohydration to accrue during exercise, will result in an increase in subjective feelings of exertion, or this likely contributed to
    the reduction in exercise performance [24]. As little as a 2% reduction
    in body mass due to insufficient hydration can also result in impaired
    cognitive function, with changes in mood state and modest reductions
    in concentration, alertness and short-term memory reported [1,24]. In
    addition to the established physiological consequences of hypohydration,
    the generally unpleasant symptoms of hypohydration (e.g. dry mouth,
    thirst, headache) may directly produce a negative effect on mood state
    [2,12]. In fact, some authors maintain that dehydration-associated
    impairment of tasks with a large cognitive component is driven primarily
    by the discomfort and distraction associated with these symptoms [6].
    Data quantifying the hydration practices of regular drivers is scarce,
    but assessments of hydration status and reported beverage intakes
    among employees in a variety of workplace settings highlighted that a
    significant proportion of employees report to work exhibiting signs of
    dehydration [25]. A large proportion of those individuals also remained
    in a state of hypohydration at the end of their shift, citing restrictions on
    when and where they could consume fluid and access to toilet facilities
    as the primary barriers to increasing water intake. It is likely that driving
    in a hot car will lead to significant losses of water over the course of a
    long journey, but these data are not readily available in the scientific
    literature. Even in an air-conditioned car, evaporative water losses
    from the skin and lungs are likely to accumulate during a long drive
    due to exposure to dry air because of the increased vapour pressure gradient. Taking these points into consideration, the European Hydration
    Institute recommends the regular ingestion of non-alcoholic beverages
    during long automobile journeys to help to reduce road fatigue [10].
    These guidelines are likely to be sound, but anecdotal reports suggest
    that many drivers avoid drinking adequately, with a view to limiting
    the need for bathroom stops during long journeys.
    While it is widely acknowledged that the use of alcohol or drugs
    among drivers increases the risk [29] and the severity [3] of road traffic
    accidents, there are currently no scientific evidence linking dehydration
    to an increased incidence of traffic accidents. At present only one recent
    study has investigated the possible effects of dehydration on simulated
    driving performance [20]. Again the primary focus was to examine the
    effects of moderate quantities of alcohol on aspects of driving performance, but this group also suggested a possible interaction between
    alcohol consumption and dehydration. The authors suggested that
    alcohol-induced impairments in cognition, and consequently on simulated driving performance, would be greater when individuals were
    also in a state of dehydration. Although the results of this study failed
    to identify any significant impact of hydration status on driving performance, it is worth noting that the simulated driving task employed was
    short (15 min) and was set in a suburban environment.
    An estimated 1.2 million people worldwide are killed as a result of
    road traffic accidents each year, with around 50 million people also injured annually [40]. Driver error is by far the largest cause of these accidents, accounting for approximately 68% of all vehicle crashes in the UK
    [7,8]. Factors including failing to look properly, misjudging another
    driver's path or speed and driver distraction are cited in the top ten
    most common causes of traffic accidents [7,8]. During long and monotonous driving, most drivers progressively show signs of visual fatigue
    and loss of vigilance [4]. Hypohydration has been shown to result in
    altered mood and deficits in aspects of cognition, it is reasonable to
    assume that dehydrated drivers may be more susceptible to errors in
    judgement and/or the successful execution of motor skill. With this in
    mind, the aim of the present study was an initial exploration of the
    effects of mild hypohydration, on performance during a prolonged, monotonous driving task where aspects of cognition relevant to driving
    (e.g. response times and loss of vigilance) are likely to be challenged.
  14. Methods
    2.1. Participants
    Twelve healthy males were recruited to participate in this randomised
    crossover design study. All participants were experienced drivers; having
    driven for over 2 years on a full licence and for more than 2 h/week. Prior
    to volunteering, participants received written information regarding the
    nature and purpose of the study and a written statement of consent
    was signed. One participant completed all trials but was excluded from
    the final results after displaying a high propensity to fall asleep during
    the driving task (perhaps caused by sleep deprivation). Physical characteristics (Mean ± SD) of the remaining 11 participants were: age 22 ±
    4 y; height 1.75 ± 0.06 m; and body mass 77.4 ± 10.0 kg. This study
    was approved by the local Ethical Advisory Committee (REF: R14-P12).
    2.2. Experimental design
    Each volunteer visited our laboratories on three separate occasions.
    The first visit was a familiarisation trial that involved the completion
    of the same driving task undertaken in the experimental trials. This
    was intended to enable the participants to become accustomed with
    the study protocol and limit any possible learning effect apparent with
    the use of the driving simulator. This was followed by two experimental
    trials. All trials were separated by at least 7 days and experimental trials
    were completed in a randomised order. Participants were provided
    with a customised diary to record dietary intake and physical activity
    during the 24 h before the first experimental trial and were asked to
    replicate this on the day prior to the subsequent experimental trials.
    During each trial period (as illustrated in Fig. 1), participants were
    asked to record dietary intake in a food and beverage record diary,
    using the portion size method. No restrictions on routine or food/
    beverage intake, other than those mentioned below, were enforced
    during this period, as the aim was to mimic free-living conditions. To
    help ensure the volunteers were adequately hydrated, they were
    instructed to consume at least 2.5 L of fluid, spread evenly across the
    day [9]. No strenuous exercise or alcohol consumption was permitted
    in the 24 h before, as well as during, each trial.
    2.3. Experimental protocol
    Each experimental trial took place over two days, as illustrated in
    Fig. 1. On day 1, volunteers visited the laboratory in the morning after
    an overnight fast (10 h, with no food or fluid permitted). A urine sample
    was obtained and body mass measured to the nearest 10 g in minimal
    clothing (underwear). Volunteers then sat for 15 min, before a 5 mL
    blood sample was collected from a superficial antecubital vein. During
    the 15 min of seated rest, subjective feelings related to thirst, hunger,
    concentration and alertness were assessed using a series of 100 mm visual analogue scales [36]. Volunteers were then free to leave the laboratory with the instruction to replicate their food intake of the pre-trial
    standardisation day. During the hydrated (HYD) trial volunteers continued to consume at least 2.5 L of fluid, spread evenly across the day.
    During the fluid restriction (FR) trial, only 25% of the HYD fluid intake
    was permitted; this was expected to result in a ~1% reduction in body
    mass over a 24 h period [36].
    Participants then returned to the laboratory the following morning
    after an overnight fast (10 h, with no food or fluid permitted). A urine
    sample was obtained and body mass measured in minimal clothing.
    314 P. Watson et al. / Physiology & Behavior 147 (2015) 313–318
    After sitting for 15 min, a 5 mL blood sample was then collected from a
    superficial antecubital vein. The same visual analogue scales were
    also completed at this time and a heart rate telemetry band was
    positioned (Polar RS400, Kempele, Finland). Participants were then provided with a standardised dry breakfast (2 cereal bars; Alpen, Weetabix
    Ltd., Kettering, UK), providing 1052 kJ, 42 g of carbohydrates, 7.6 g of fat
    and 3.8 g of protein. During the HYD trial they were given a volume of
    plain water to drink with breakfast (500 mL), but on the FR trial only
    a very small volume was provided (50 mL). They were then fitted
    with electroencephalogram (EEG) and electrooculogram (EOG) electrodes. Electrodes were attached for two channels of EEG, with interelectrode distances carefully maintained by using the ‘10–20 EEG montage’ (main channel C3–A1, backup channel C4–A2), and there were
    two EOG channels (electrodes 1 cm lateral to and below left outer canthus and 1 cm lateral to and above right outer canthus; both referred to
    the centre of the forehead).
    One hour following breakfast, volunteers began a driving simulation
    task, similar to that described in several publications [11,30,31]. The
    task comprised of a 2 h continuous drive in an immobile car with a
    full-size, interactive, computer-generated road projection of a dull monotonous dual carriageway, each carriageway having two lanes. The
    road also had a hard shoulder and simulated auditory ‘rumble strips’
    (incorporated into white lane markings) either side of the carriageway
    and a barrier separating the carriageways, with long straight sections
    followed by gradual bends. Slow moving vehicles were met occasionally, and these had to be overtaken. Drivers were instructed to remain
    within their lane unless overtaking. During the HYD trial volunteers
    were be provided with 200 mL of fluid every hour, and on the FR trial
    only 25 mL was made available each hour. Immediately following the
    drive, volunteers then sat for 15 min, before a 5 mL blood sample was
    collected from a superficial antecubital vein. A final assessment of subjective feelings related to thirst, throat dryness, hunger, concentration
    and alertness was undertaken, before a urine sample was obtained
    and body mass was again measured in minimal clothing.
    2.4. Analysis
    2.4.1. Dietary intake
    Nutritional analysis of food intake records was undertaken using
    commercially-available nutritional analysis software (NetWISP v4.0,
    Tinuviel Software, UK). Total water intake from all food and drink, as
    determined from food composition tables within the database, was
    the primary focus. The contribution of metabolic water to total body
    water was not accounted for, as this was assumed to be consistent
    across both trials. Energy, macronutrient and caffeine intakes were
    also examined to ensure consistency across trials.
    2.4.2. Driving related measures
    Instances of lane drifting or late breaking are the most common
    manifestation of driver error, and a car wheel touching (or crossing)
    the rumble strip or lane line was identified as a driving ‘incident’.
    These were classified as ‘minor incidents’, whereas ‘major incidents’ included cases where the car completely leaves the lane, hits the barrier or
    another car. Split-screen video footage of the roadway and driver's face
    (filmed by an unobtrusive infrared camera) enables the cause of the
    incident to be determined. Those due to sleepiness (e.g. excessive
    blinking, eye closure, eyes rolling upwards or vacant staring ahead)
    were logged as ‘sleep-related incidents’. Non-sleep related incidents
    (driver distraction, fidgeting or looking around) are also recorded.
    2.4.3. EEG and EOG
    EEGs and EOGs were recorded using “Embla” (Flaga Medica Devices,
    Iceland) and spectrally analysed using “Somnologica” (Flaga) in 4 s
    epochs. EEG low and high band-pass filtering at N20 Hz removed slow
    eye movements and muscle artefacts. Increases in EEG power in the
    alpha (8–11 Hz) and theta (4–7 Hz) ranges indicate increasing sleepiness
    [22] and reduced vigilance [4]. EEG power in this (4–11 Hz) frequency
    range was then averaged in one-minute epochs. To remove individual
    differences in these EEG power levels and to permit better comparison
    between conditions, these data were standardised for each participant
    by taking the difference between each epoch and mean value for that
    person's EEG power during the first 30 min of the HYD trial, divided by
    the standard deviation around that mean [11,31].
    2.4.4. Blood and urine samples
    Blood samples collected throughout the experimental protocol were
    drawn into dry syringes before being dispensed into plain tubes and left
    to clot at room temperature for 1 h. These samples were then centrifuged at 3000 g for 10 min to yield serum. When urine samples were
    obtained, participants were instructed to empty their bladder as
    completely as possible into a collection container. The volume of each
    void was determined, and a 5 mL aliquot was retained in a sterile collection tube. All urine and serum samples were stored at 4 °C for a maximum of 7 days before being analysed for osmolality using freezing
    point depression (Gonotoc Auto; Berlin, Germany).
    Fig. 1. A schematic representation of the experimental protocol describing the methods employed to manipulate of hydration status on days 1 and 2 of the trials. Arrows indicate the measurement of body mass, urine and serum osmolality undertaken upon arrival laboratory at the start of days 1 & 2 of the experimental protocol, as well as the end of the driving protocol.
    P. Watson et al. / Physiology & Behavior 147 (2015) 313–318 315
    2.5. Statistical analysis
    On the basis of the results of previous investigations undertaken
    using the same experimental model [11,18,30,31], we estimated a 90%
    probability of detecting a difference in total errors of at least 32 with a
    sample size of 11 subjects (G-Power 3.1, Dusseldorf, Germany). Data
    are presented as mean ± standard deviation (SD) unless otherwise stated. Driving incidents and the EEG data were averaged into 30 min
    epochs, as described by Reyner et al. [31]. The distribution of the data
    was first assessed using the Shapiro–Wilk test. Differences in the total
    number of driver errors recorded during each trial, as well as the baseline measures used to check pre-trial standardisation, were assessed
    using paired sample t-tests. Cohen's d effect sizes (ES) for the differences in driver error rate were also determined. To identify differences
    in normally-distributed data collected throughout each trial, two-way
    (time-by-trial) ANOVA were employed. Where a significant interaction
    was apparent, pair-wise differences were evaluated using the
    Bonferroni correction. For the purpose of hypothesis testing, the 95%
    level of confidence was predetermined as the minimum criterion to
    denote a statistical difference (P b 0.05).
  15. Results
    Pre-trial body mass (t = 0.391, P = 0.692), urine osmolality
    (t = −0.216, P = 0.838), serum osmolality (t = 0.338, P = 0.574)
    were the same on both trials, suggesting that the participants were in
    a similar state of hydration before the start of each trial. Pre-trial dietary
    energy (HYD 12.6 ± 1.2 MJ; FR 12.3 ± 1.8 MJ; t = 0.297, P = 0.742) and
    caffeine (HYD 157 ± 51 mg; FR 131 ± 46 mg; t = 0.412, P = 0.742) intakes were also not different. Participants also started both trials
    reporting the same subjective feelings of thirst, throat dryness, hunger,
    alertness and ability to concentrate; further supporting this view.
    Total water intake from all sources during day 1 of the HYD trial
    was 3.0 ± 0.2 L, compared to 0.9 ± 0.1 L ingested during the FR trial
    (t = 10.647, P b 0.001). This comprised 2.6 ± 0.2 L from beverages
    and 0.4 ± 0.2 L from foods in the HYD trial, whereas 0.5 ± 0.2 L and
    0.4 ± 0.1 L was ingested through beverages and foods respectively
    during the FR trial. Caffeine intake was lower during the FR trial
    (55 ± 12 mg) compared to the HYD trial (208 ± 49 mg; P = 0.017).
    FR during day 1 resulted in a 1.1 ± 0.7% (range −0.7 to −2.3%) reduction in body mass, compared to −0.1 ± 0.6% (range +1.1 to −0.7%) in
    the HYD trial (F = 38.482, P = 0.002). The 24 h restriction of fluid intake
    resulted in an increase in both serum (F = 92.042, P = 0.007) and urine
    osmolality (F = 207.904, P b 0.001; Fig. 2).
    The number of driver errors made during the trials, both minor and
    major incidents, grouped into 30 min blocks, is illustrated in Fig. 3. There
    was a progressive increase in the total number of driver errors observed
    during the HYD trial, with significantly more incidents recorded during
    the last 30 min period (17 ± 16), than in the first 30 min (7 ± 8; F =
    3.587, P = 0.043). However, the frequency of driver error increased to
    a greater extent throughout the FR trial (F = 8.043, P = 0.008). FR resulted in a marked increase in the total number of driving errors, with
    47 ± 44 and 101 ± 84 recorded during the HYD and FR trials respectively (t = −4.549, P = 0.006; ES = 0.81). Four major incidents were recorded over the course of the study, but these were evenly distributed
    between the HYD and FR trials. There was no clear relationship between
    the number of errors made during the FR trial and the degree of dehydration accrued (r2 = 0.18; P = 0.544); it is likely that there is insufficient statistical power to detect such an effect with the number of
    participants recruited, nor was the experiment designed to examine
    this question.
    The analysis of the EEG data is presented in Fig. 4. There was a
    progressive increase in alpha (8–11 Hz) and theta (4–7 Hz) activity
    throughout both the HYD and FR trials (F = 4.528, P = 0.038), indicative of greater sleepiness and perhaps reduced vigilance. The magnitude
    of change tended to be greater in the FR trial, but this response just
    failed to reach significance (F = 2.998, P = 0.062).
    There was no change in thirst perception over the course of the HYD
    trial, but self-reported ratings of thirst increased by 107 ± 17% throughout the FR trial (F= 80.920, P b 0.001). The same response was apparent
    when examining the perceived feelings of throat dryness. Perceived
    ability to concentrate (−39 ± 17%; F = 22.475, P b 0.001) and alertness
    (−48 ± 26%; F = 6.845, P = 0.016) had also reduced over the course of
    the FR trial, but these were both significantly lower at the end of the
    Fig. 2. Urine (top) and serum (bottom) osmolality throughout the HYD and FR trials. *
    denotes a significant difference between trials at the corresponding time point
    (P b 0.05). Data are presented as mean ± standard deviation.
    Fig. 3. The total number of driver errors made during each 30 min period of the HYD and
    FR trials. * denotes a significant difference between trials at the corresponding time point
    (P b 0.05). Data are presented as mean ± standard deviation.
    316 P. Watson et al. / Physiology & Behavior 147 (2015) 313–318
    drive during the FR trial than compared with the HYD trial (both
    P b 0.001).
  16. Discussion
    4.1. General discussion
    Driver error is by far the largest cause of road traffic accidents, accounting for approximately 68% of all vehicle crashes in the UK [7,8].
    During motorway/highway driving, drivers tend to progressively
    show signs of visual fatigue and loss of vigilance [4]. Since deficits in
    TBW are associated with altered mood and decrements in aspects of
    cognitive function, it is possible that dehydrated drivers may be prone
    to making more errors in judgement and car handling. The results of
    this exploratory study suggest that mild hypohydration, induced
    through a short-term period of fluid restriction, produced in a significant increase in minor driving errors during a prolonged, monotonous
    drive, compared to that observed while performing the same task in a
    hydrated condition. Mild dehydration can produce negative changes
    in mood state and modest reductions in concentration, alertness and
    short-term memory [1,2,12,14]. While there remains some uncertainty
    whether these responses result from a physiological impairment caused
    by the reduction in total body water and electrolyte imbalance, or are
    simply due to the discomfort and distraction associated with dehydration, these subtle changes in mood and cognition mostly likely explain
    the decrement in driving performance observed.
    Water accounts for 50–60% of body mass in most healthy individuals, and maintaining water balance is essential for health. Body water
    turnover rate, a function of fluid losses (respiratory water, sweat,
    urine, faeces) and fluid gain from food, beverages and metabolic
    water, is highly variable between individuals with typical values of between three to six litres/day reported in the literature. The homeostatic
    regulation of salt and water balance normally acts to limit excursions in
    total body water to no more than about 1% per day [5]. Exposure to environmental extremes (particularly heat and humidity) and prolonged
    physical activity, as well as some nutritional (fluid restriction, alcohol)
    and pharmacological (diuretics) interventions can significantly accelerate fluid losses over time. Hypohydration causes a reduction in the
    circulating blood volume, a reduction in stroke volume and an elevated
    heart rate at a given exercise intensity [13,26]. There is also evidence of
    direct effects of hypohydration on the central nervous system [28,39],
    which may contribute to these observed changes in both mood and
    cognitive function.
    The American College of Sports Medicine qualifies mild hypohydration as body mass losses exceeding 1%, and as such are deviations
    in total body water that may be encountered routinely by adults during
    daily activities [33]. While data quantifying the hydration practices of
    regular drivers is scarce, when hydration status has been assessed in a
    variety of workplace settings, a significant proportion of employees report to work exhibiting signs of dehydration [25]. In addition, it is likely
    that driving in a hot car may lead to significant losses of water over the
    course of a long journey, but again these data are not readily available in
    the scientific literature. While it has been suggested that drivers should
    aim to regularly ingest non-alcoholic beverages during long automobile
    journeys to help to reduce road fatigue [10], and caffeine containing
    beverages, including coffee and energy drinks, are regularly promoted
    to counteract driver fatigue [30], factors such as limited free access to
    fluids and desire to avoid stops for bathroom breaks mean that drivers
    may place themselves at greater risk of dehydration.
    At present only one study has examined the effects of dehydration
    on simulated driving performance [20]. This particular study was primarily designed to investigate a possible interaction between
    exercise-induced dehydration and alcohol consumption, as the authors
    suggest that many people tend to consume alcoholic beverages following participation in sports. No effect of dehydration was observed, but it
    is worth noting that the driving task employed was particularly short
    (15 min). While several studies have reported decrements in aspects
    of cognitive function with dehydration [1,2,12,14], there are a number
    of conflicting reports suggesting little or no change in cognition following a variety of dehydration protocols [21,37,38]. Innate intelligence and
    life experience of familiar day-to-day tasks, such as driving, result in
    functionally more efficient cognitive networks and therefore provide a
    cognitive reserve [34]. This acts as a buffer providing resilience to cope
    with increasingly complex tasks while still functioning adequately,
    and also delays the onset of clinical manifestations of neurodegenerative disorders such as Alzheimer's disease. There is evidence that
    individuals are able to tolerate a degree of dehydration without any
    measureable impairment in cognition by increasing the degree of
    brain activation required for a given task [21]. It appears likely that in
    the present study the task was sufficiently complex and long lasting to
    overcome this reserve capacity and result in a measurable decrement
    in cognitive performance. It is worth noting that while some studies
    do not report significant differences in task performance with varying
    levels of dehydration, these data suggest that losses of body water
    and/or electrolyte imbalances do appear to produce decrements in aspects of brain function underlying important cognitive processes.
    4.2. Implications and limitations of the study
    Driving performance in the present study was assessed through a
    simulated driving task, rather than ‘real world’ on-road driving. The
    use of a driving simulator allowed us to study long, monotonous, and
    uninterrupted driving task, but it is difficult to know how to translate
    the driving errors measured during a simulated drive to the likelihood
    of accidents occurring on the road. The simulator employed in the
    present study has been internally validated against an instrumented
    real car circulating a race track, and it has been employed in several published studies investigating the link between tiredness and driving performance publications [11,18,30,31]. The car cabin environment,
    including commands and instruments were identical to an operative
    car, but it should be recognised that a driving simulation is not real driving. While participants were instructed to drive as diligently as they
    would on the road, the consequences of a minor error made during a
    simulation are clearly not the same as would be experienced while driving at speed on a motorway [4]. However, the present data do suggest
    that decrements in vigilance, decision making and mood, as apparent
    from the EEG data and the reported subjective feelings, are likely to
    have a significant influence on driving behaviour. This is likely to translate into a greater potential for errors in both simulated and real world
    Fig. 4. EEG alpha + theta power (4–11 Hz) averaged every 30 min and normalised against
    each individual's power in these ranges, by taking the difference between each minute's
    epoch and the individual's mean value over the first 30 min of the HYD data, and dividing
    this by the standard deviation around the mean of that 30 min of data. Data are presented
    as mean ± standard deviation.
    P. Watson et al. / Physiology & Behavior 147 (2015) 313–318 317
    settings, and consequently influencing the possibility of road traffic
    accidents.
    Driver fatigue and sleep-related vehicle accidents account for a considerable proportion of all vehicle accidents, especially those on motorways and other monotonous roads [17]. These types of accidents are of
    particular concern since the possibility of a fatality is approximately
    three times greater than encountered in general road accidents [7,8].
    Many road traffic accidents are preventable, and a variety of national
    initiatives and targets to reduce road deaths and serious injuries have
    been implemented in recent years [7,8]. Interventions that have been
    implemented both within the UK and elsewhere to prevent or reduce
    the occurrence of accidents on the road and the severity of injuries
    sustained, include changes to the road environment, media safe driving
    campaigns, drink driving campaigns, stricter enforcement of legislation
    relating to roads, and finally targeted driver education programmes. The
    later approach aims to enhance safe driving skills through increased
    awareness of the dangers involved and improved recognition of driving
    hazards. While the effects of alcohol consumption and driving while
    tired is mentioned in these courses, there is no mention of other factors
    that drivers should be aware of to maintain attention and vigilance.
    In conclusion, the results of this initial exploratory study suggest that
    mild dehydration, induced through a short-term period of fluid restriction, produced in a significant increase in minor driving errors during a
    prolonged, monotonous drive, compared to that observed while
    performing the same task in a hydrated condition. Due to the nature
    of the experimental protocol, it is unclear whether this response was
    caused by prior the fluid restriction, difference in fluid intake during
    the drive or combination of both these factors. Further work is warranted to examine contribution of these factors to the response observed.
    The level of dehydration induced in the present study was mild and
    could easily be reproduced by individuals with limited access to fluid
    over the course of a busy working day. To provide some context to the
    magnitude of decrement in stimulator performance reported, a similar
    increase in driver error rate has been observed when driving following
    the ingestion of an alcoholic beverage resulting in a blood alcohol content of approximately 0.08% (the current UK legal driving limit), or
    while sleep deprived [19]. There is no question that both drink-driving
    and driving while tired increases the risk of road traffic accidents [40],
    and many countries have instigated national campaigns to educate
    drivers of the associated risks. Given the present findings, perhaps
    some attention should also be directed to encouraging appropriate hydration practices among drivers.
    Acknowledgements
    This work was funded in part by a grant from the European Hydration
    Institute (EHI). The EHI did not directly contribute to the study design;
    the collection, analysis and interpretation of data or in the writing of
    the manuscript.
    References
    [1] A. Adan, Cognitive performance and dehydration, J. Am. Coll. Nutr. 31 (2012) 71–78.
    [2] L.E. Armstrong, M.S. Ganio, D.J. Casa, E.C. Lee, B.P. McDermott, J.F. Klau, L. Jimenez, L.
    Le Bellego, E. Chevillotte, H.R. Lieberman, Mild dehydration affects mood in healthy
    young women, J. Nutr. 142 (2012) 382–388.
    [3] A. Bradbury, Pattern and severity of injury sustained by pedestrians in road traffic
    accidents with particular reference to the effect of alcohol, Injury 22 (1991)
    132–134.
    [4] A. Campagne, T. Pebayle, A. Muzet, Correlation between driving errors and vigilance
    level:influence of the driver's age, Physiol. Behav. 80 (2004) 515–524.
    [5] S.N. Cheuvront, R. Carter, S.J. Montain, M.N. Sawka, Daily body mass variability and
    stability in active men undergoing exercise-heat stress, Int. J. Sport Nutr. Exerc.
    Metab. 14 (2004) 532–540.
    [6] S.N. Cheuvront, R.W. Kenefick, Dehydration: physiology, assessment, and performance effects, Comp. Physiol. 4 (2014) 257–285.
    [7] Department of Transport, Reported Road Casualties in Great Britain: 2011 Annual
    Report, https://www.gov.uk/government/uploads/system/uploads/attachment_
    data/file/9274/rrcgb2011-01.pdf 2012 (Accessed on 1st September 2014).
    [8] Department of Transport (2012b) A safer way. Consultation on making Britain's
    roads the safest in the world. http://www.dft.gov.uk/consultations/archive/2009/
    roadsafetyconsultation/roadsafetyconsultation.pdf (Accessed on 1st September 2014).
    [9] EFSA Panel on Dietetic Products, Nutrition, and Allergies, Scientific opinion on
    dietary reference values for water, EFSA J. 8 (3) (2010) 1459.
    [10] European Hydration Institute, 4 key moments in hydration, http://www.
    europeanhydrationinstitute.org/4_key_moments_in_hydration.html 2014 (accessed
    1st September 2014).
    [11] A.J. Filtness, L.A. Reyner, J.A. Horne, Driver sleepiness-comparisons between young
    and older men during a monotonous afternoon simulated drive, Biol. Psychol. 89
    (2012) 580–583.
    [12] M.S. Ganio, L.E. Armstrong, D.J. Casa, B.P. McDermott, E.C. Lee, L.M. Yamamoto, S.
    Marzano, R.M. Lopez, L. Jimenez, L. Le Bellego, E. Chevillotte, H.R. Lieberman, Mild
    dehydration impairs cognitive performance and mood of men, Br. J. Nutr. 106
    (2011) 1535–1543.
    [13] J. González-Alonso, R. Mora-Rodríguez, P.R. Below, E.F. Coyle, Dehydration reduces
    cardiac output and increases systemic and cutaneous vascular resistance during
    exercise, J. Appl. Physiol. 79 (1995) 1487–1496.
    [14] P.M. Gopinathan, G. Pichan, V.M. Sharma, Role of dehydration in heat stress-induced
    variations in mental performance, Arch. Environ. Health 43 (1988) 15–17.
    [17] J.A. Horne, L.A. Reyner, Sleep related vehicle accidents, BMJ 310 (1995) 565–567.
    [18] J.A. Horne, L.A. Reyner, Counter-acting driver sleepiness: effects of napping, caffeine,
    and placebo, Psychophysiology 33 (1996) 306–309.
    [19] J.A. Horne, L.A. Reyner, P.R. Barrett, Driving impairment due to sleepiness is
    exacerbated by low alcohol intake, Occup. Environ. Med. 60 (2003) 689–692.
    [20] C. Irwin, M. Leveritt, D.H. Shum, B. Desbrow, Mild to moderate dehydration
    combined with moderate alcohol consumption has no influence on simulated
    driving performance, Traffic Inj. Prev. 15 (2014) 652–662.
    [21] M.J. Kempton, U. Ettinger, R. Foster, S.C. Williams, G.A. Calvert, A. Hampshire, F.O.
    Zelaya, R.L. O'Gorman, T. McMorris, A.M. Owen, M.S. Smith, Dehydration affects brain
    structure and function in healthy adolescents, Hum. Brain Mapp. 32 (2011) 71–79.
    [22] T. Kerstedt, M. Gillberg, Subjective and objective sleepiness in the active individual,
    Int. J. Neurosci. 52 (1990) 29–37.
    [24] R.J. Maughan, Impact of mild dehydration on wellness and on exercise performance,
    Eur. J. Clin. Nutr. 57 (2003) S19–S23.
    [25] S.A. Mears, S.M. Shirreffs, Assessing hydration status and reported beverage intake in the workplace, Am. J. Lifestyle Med. (2014)http://dx.doi.org/10.1177/
    1559827614523706.
    [26] S.J. Montain, E.F. Coyle, Influence of graded dehydration on hyperthermia and
    cardiovascular drift during exercise, J. Appl. Physiol. 73 (1992) 1340–1350.
    [28] L. Nybo, N.H. Secher, Cerebral perturbations provoked by prolonged exercise, Prog.
    Neurobiol. 72 (2004) 223–261.
    [29] E.J. Ogden, H. Moskowitz, Effects of alcohol and other drugs on driver performance,
    Traffic Inj. Prev. 5 (2004) 185–198.
    [30] L.A. Reyner, J.A. Horne, Efficacy of a ‘functional energy drink’ in counteracting driver
    sleepiness, Physiol. Behav. 75 (2002) 331–335.
    [31] L.A. Reyner, S.J. Wells, V. Mortlock, J.A. Horne, ‘Post-lunch’ sleepiness during
    prolonged, monotonous driving - effects of meal size, Physiol. Behav. 105 (2012)
    1088–1091.
    [32] M.N. Sawka, A. Young, R.P. Francesconi, S.R. Muza, K.B. Pandolf, Thermoregulatory
    and blood responses during exercise at graded hypohydration levels, J. Appl. Physiol.
    59 (1985) 1394–1401.
    [33] M.N. Sawka, L.M. Burke, et al., American college of sports medicine position stand.
    Exercise and fluid replacement, Med. Sci. Sports Exerc. 39 (2007) 377–390.
    [34] N. Scarmeas, Y. Stern, Cognitive reserve and lifestyle, J. Clin. Exp. Neuropsychol. 25
    (2003) 625–633.
    [36] S.M. Shirreffs, S.J. Merson, S.M. Fraser, D.T. Archer, The effects of fluid restriction on
    hydration status and subjective feelings in man, Br. J. Nutr. 91 (2004) 951–958.
    [37] G. Szinnai, H. Schachinger, M.J. Arnaud, L. Linder, U. Keller, Effect of water deprivation on cognitive-motor performance in healthy men and women, Am. J. Physiol.
    Regul. Integr. Comp. Physiol. 289 (2005) R275–R280.
    [38] P.D. Tomporowski, K. Beasman, M.S. Ganio, K. Cureton, Effects of dehydration and
    fluid ingestion on cognition, Int. J. Sports Med. 28 (2007) 891–896.
    [39] P. Watson, K.E. Black, S.C. Clark, R.J. Maughan, Exercise in the heat: effect of fluid
    ingestion on blood–brain barrier permeability, Med. Sci. Sports Exerc. 38 (2006)
    2118–2124.
    [40] World Health Organization. (2004) World report on road traffic injury prevention.
    Geneva: World Health Organization. http://whqlibdoc.who.int/publications/2004/
    9241562609.pdf (accessed on 1st September 2014).
    318 P. Watson et al. / Physiology & Behavior 147 (2015) 313–318
    ???
    C BY NC ND
    A Rough Guide to
    SPOTTING BAD SCIENCE
    © COMPOUND INTEREST 2015 - WWW.COMPOUNDCHEM.COM | @COMPOUNDCHEM
    Shared under a Creative Commons Attribution-NonCommercial-NoDerivatives licence.
  17. SENSATIONALISED HEADLINES
    Aa
    Article headlines are commonly designed to
    entice viewers into clicking on and reading
    the article. At times, they can over-simplify
    the fndings of scientifc research. At worst,
    they sensationalise and misrepresent them.
  18. MISINTERPRETED RESULTS
    News articles can distort or misinterpret the
    fndings of research for the sake of a good
    story, whether intentionally or otherwise. If
    possible, try to read the original research,
    rather than relying on the article based on
    it for information.
  19. CONFLICTS OF INTEREST
    Many companies will employ scientists to
    carry out and publish research - whilst this
    doesn’t necessarily invalidate the research,
    it should be analysed with this in mind.
    Research can also be misrepresented for
    personal or fnancial gain.
  20. CORRELATION & CAUSATION
    Be wary of any confusion of correlation and
    causation. A correlation between variables
    doesn’t always mean one causes the other.
    Global warming increased since the 1800s,
    and pirate numbers decreased, but lack of
    pirates doesn’t cause global warming.
  21. UNSUPPORTED CONCLUSIONS
    Speculation can often help to drive science
    forward. However, studies should be clear
    on the facts their study proves, and which
    conclusions are as yet unsupported ones. A
    statement framed by speculative language
    may require further evidence to confrm.
  22. PROBLEMS WITH SAMPLE SIZE
    In trials, the smaller a sample size, the
    lower the confdence in the results from
    that sample. Conclusions drawn can still be
    valid, and in some cases small samples are
    unavoidable, but larger samples often give
    more representative results.
  23. UNREPRESENTATIVE SAMPLES USED
    In human trials, subjects are selected that
    are representative of a larger population. If
    the sample is diferent from the population
    as a whole, then the conclusions from the
    trial may be biased towards a particular
    outcome.
  24. NO CONTROL GROUP USED
    In clinical trials, results from test subjects
    should be compared to a ‘control group’ not
    given the substance being tested. Groups
    should also be allocated randomly. In
    general experiments, a control test should
    be used where all variables are controlled.
  25. NO BLIND TESTING USED
    To try and prevent bias, subjects should
    not know if they are in the test or the
    control group. In ‘double blind’ testing,
    even researchers don’t know which group
    subjects are in until after testing. Note,
    blind testing isn’t always feasible, or ethical.
  26. SELECTIVE REPORTING OF DATA
    Also known as ‘cherry picking’, this involves
    selecting data from results which supports
    the conclusion of the research, whilst
    ignoring those that do not. If a research
    paper draws conclusions from a selection
    of its results, not all, it may be guilty of this.
  27. UNREPLICABLE RESULTS
    Results should be replicable by independent
    research, and tested over a wide range of
    conditions (where possible) to ensure they
    are consistent. Extraordinary claims require
    extraordinary evidence - that is, much more
    than one independent study!
  28. NON-PEER REVIEWED MATERIAL
    Peer review is an important part of the
    scientifc process. Other scientists appraise
    and critique studies, before publication
    in a journal. Research that has not gone
    through this process is not as reputable,
    and may be fawed.
    x x
    Being able to evaluate the evidence behind a scientifc claim is important. Being able to recognise bad science reporting, or
    faults in scientifc studies, is equally important. These 12 points will help you separate the science from the pseudoscience.