Sex volume

Not a free member yet?

Currently known as: The Journal of Sex Research ( - current). Incorporates. Annual Review of Sex Research ( - ). List of issues. Volume 56 Sex Volume 5 Paperback – February 21, ​ by Joe Casey (Author), Piotr Kowalski (Artist), Sonia Harris (Artist) & 0 more.​ Sex Volume 4: Daisy Chains by Joe Casey Paperback $ In this study, we aimed to investigate the effect of sex × BMI interactions on gray matter volume (GMV), and to determine the implications of any.

Sax For Sex - Volume 3, an album by The Rainbow Orchestra on Spotify. Sex Volume 5 Paperback – February 21, ​ by Joe Casey (Author), Piotr Kowalski (Artist), Sonia Harris (Artist) & 0 more.​ Sex Volume 4: Daisy Chains by Joe Casey Paperback $ Studies assessing volumetric sex differences have provided contradictory results. Total intracranial volume (TIV) is a major confounding factor.

In this study, we aimed to investigate the effect of sex × BMI interactions on gray matter volume (GMV), and to determine the implications of any. Currently known as: The Journal of Sex Research ( - current). Incorporates. Annual Review of Sex Research ( - ). List of issues. Volume 56 Sex Volume 5 Paperback – February 21, ​ by Joe Casey (Author), Piotr Kowalski (Artist), Sonia Harris (Artist) & 0 more.​ Sex Volume 4: Daisy Chains by Joe Casey Paperback $






Metrics details. Studies assessing volumetric sex differences have provided sex results. We investigated how the number, size, and direction of sex differences in gray matter volume GMv vary depending on how TIV variation is statistically handled.

Sex differences in the GMv of VOIs were assessed in participants females without correcting for TIV variation or after adjusting the data with 5 different methods Volume non-linear-only modulation, proportions, power-corrected-proportions, covariation, and the residuals method.

Linear regression was used to quantify TIV effects on raw GMv and the efficacy of each method in controlling for them. All TIV-adjustment methods reduced the number of sex differences but their results were very different. There is not just one answer to the question of how many and how large the sex differences in GMv are, but not all the possible answers are equally valid. When TIV effects are ruled out using appropriate adjustment methods, few sex differences if any remain statistically significant, and their size is quite reduced.

The subject of neuroanatomical sex differences in the brain is unique due to its scientific importance [ 1234 ] and social repercussions [ 56 ]. However, precisely quantifying sex differences in the volumes of specific brain regions is a challenging task, and studies assessing volumetric sex differences have provided heterogeneous and inconsistent results.

The same thing occurs for many other gray and white matter structures e. The inconsistencies and contradictions in the results of different studies evaluating volumetric sex differences are probably caused by many factors. However, it is believed that one of the major difficulties in these kinds of studies is that males and females differ in overall body and head size [ 1112151617 ]. In other words, because sex differences in gross morphology may affect global and regional brain volumes, these differences introduce a major allometric challenge that might be subdivided into three hierarchically organized methodological questions.

First, the decision has to be made whether or not to adjust raw neuroanatomical volumes. This decision is quite important because unadjusted measures seem to affect the number and direction of sex differences in brain regional volumes [ 8911121316171819 ]. Nevertheless, there seem to be pros and cons of using both raw and adjusted volumetric measurements.

Thus, adjusted brain measures are less reliable than unadjusted ones [ 20 ], but adjusted measures are currently considered more valid [ 212223 ]. A second methodological decision refers to which variable should be chosen to adjust the gross morphological variations associated with sex.

Several measures have been used for this purpose, including body weight, height, head circumference, total intracranial volume TIVand total brain volume. However, although they are still used by some researchers [ 24volume26 ], body size parameters such as height volume weight show weak and inconsistent correlations with overall brain size [ 2728 ], and they are generally perceived as inappropriate.

The inadequacy of body size parameters as possible adjustment factors would be aggravated when trying to assess small regional volumes; therefore, total volume volume and TIV are usually preferred for a more detailed discussion on this topic, see [ 29 ].

Finally, after having decided to adjust their data and which adjustment factor to use e. Three methods proportions, residuals, and covariate have frequently been used to correct TIV scaling effects [ 30 ].

Two recent studies [ 1617 ] were specifically devoted to assessing whether the use of each of these adjustment methods affects the number and direction of brain volumetric sex differences.

These studies showed that the use of proportion-adjusted data results in a larger number of sex differences, often indicating larger proportional gray matter volumes in females. By contrast, when using either of the other two methods, the number of sex differences is reduced, and their direction varies depending on the neuroanatomical region being considered.

Therefore, evidence provided by these and other studies e. Therefore, the present study was designed to confirm and extend the results of these studies by evaluating the results of five different TIV-adjustment methods in the brain areas defined by the Automated Anatomical Labeling atlas AAL [ 33 ].

More specifically, the aim of this study was fourfold. First, we aimed to assess to what extent sex differences in raw gray matter volumes are driven by TIV scaling effects. Third, we tried to validate these methods by assessing A which of them satisfactorily removed TIV-scaling effects and B how their results compared to each other and to those obtained in three criterial subsamples.

Fourth, we tried to summarize the most reliable differences by integrating the results obtained with the adjustment methods that were found to remove TIV effects. We would like to note that the present study focuses on the statistical description of the possible female-male differences in gray matter volume but it does not assess whether or not they might have functional sex behavioral consequences.

However, the use of this term does not imply any assumption on the possible origin of the observed differences a topic that was not explored in the present manuscript, either.

For this study, we collected the scans of healthy subjects females; males who had participated in previous studies by our research team, recruited through local advertisements and word of mouth. All participants were physically and psychologically healthy, with no history of neurological or psychiatric disorders. In short, male participants were slightly older than female participants M Therefore, age was not considered a relevant variable in this study.

The majority of participants Consequently, educational level was not considered a relevant variable in the present study. A total of 74 pairs of TIV-matched participants were created, resulting in two highly similar groups and a total subsample of subjects. To create these groups, participants of each sex were sorted in ascending order by their TIVs and median split into two equally sized participant pools.

Seventy-four participants were first randomly selected from each participant pool, and the sex in the TIV averages of the resulting groups was calculated. This made it possible to ascertain whether sex TIV or the sex factor was able to produce a larger number of differences, and which of them mediated most in the differences observed in the main sample.

MRI data were collected on a 1. CAT12 preprocessing was conducted following the standard default procedure suggested in the manual. Images were not smoothed because we were only interested in the modulated images. Note that this procedure does not include any correction for overall head size e. In addition, also following the standard CAT12 procedure, the total intracranial volume TIV was calculated as the sum of the gray matter, white matter, and cerebrospinal fluid volumes obtained in the tissue class images in native space.

Until the volume development of the CAT12 software, VBM8 was probably one of the most popular programs for analyzing structural neuroimaging data. Similarly to the CAT12, this protocol includes five main steps: 1 segmentation of the images into gray matter, white matter, and cerebrospinal fluid; 2 registration to a standard template provided by the International Consortium of Brain Mapping ICBM ; 3 a high-dimensional DARTEL normalization of the gray matter segments to the MNI template; 4 non-linear modulation a step in which the normalized gray matter segments are multiplied only by the non-linear determinants of the normalization deformation matrix to correct the images for individual differences in size [ 32 ]; and 5 data quality check in which no outliers or incorrectly aligned cases were detected.

To isolate the effects of the TIV-adjustment introduced by the non-linear modulation step and ensure that the outcomes of the VBM8-adjusted dataset were fully comparable to those of all the other adjustment methods, a second set of VBM8 images was obtained. This method implicitly assumes a proportional relationship between TIV and the volume of any neuroanatomical structure of interest VOI.

The adjusted volume VOI adj is individually calculated according to the following formula:. Volume, it allows estimating the group in this case, sex effects without any volume of the TIV effect, by simultaneously introducing TIV and sex as putative predictors of each VOI in a multiple regression model, resulting in the following formula:. This method incorporates information from all the participants, and having a similar number of participants in each group sex becomes critical to ensure the reliability of the results [ 16 ].

Moreover, each regression coefficient is associated with a significance level, thus making second-level analyses i. The power-corrected proportion method PCP was recently proposed by Liu et al. This method explicitly assumes that the relationship between the TIV and a VOI is not proportional, but instead follows a power law.

This procedure was initially discussed by Arndt et al. The significance threshold was initially set at 0. To maximize statistical power, no corrections for multiple comparisons were initially introduced, and following recent recommendations of the American Statistics Association [ 3637 ], we focused our analysis on effect sizes rather than p values.

More specifically, in decreasing order according to their expected statistical power, the Benjamini, Krieger and Yekutieli [ 38 ] Benjamini and Hochberg [ 39 ], Holm [ 40 ] and Bonferroni-Dunn [ 41 ] corrections for multiple comparisons were tested.

The percent of overlap denotes the proportion of scores that overlap in two normal distributions which means differ in some magnitude, whereas the percent of superiority denotes the probability that a randomly sampled member of population a will have a score Y a that is higher than the score Y b attained by a randomly sampled member from population b [ 46 ].

Previous studies have shown that in the absence of any correction, the local volumes of particular brain areas are directly related to the TIV [ 15171829 ]. The presence of this relationship in our own raw data was assessed by performing linear regression analyses relating the TIV and volume of the VOIs considered in this study. TIV-VOI adj relationships provided a first and powerful criterion to evaluate the goodness of the different adjustment methods tested in this study. That is, because the aim of the adjustment methods is to get rid of TIV effects and provide an unadulterated estimation of sex differences, satisfactorily adjusted data should not show the linear TIV-VOI adj relationship predicted for the raw data, and the likelihood or size of sex differences in local gray matter volumes should not sex associated with TIV-VOI adj slope values.

In these analyses, p values were used instead of test statistics because the former provide standardized versions of the latter that can be compared across all the adjustment methods and samples used in the present study for a more detailed discussion, sex [ 30 ]. To obtain a more detailed comparison with the TIV-matched subsample, we analyzed the relative frequency of coincidental and non-coincidental findings of this criterial subsample and each TIV-adjusted dataset.

A coincidental result hit was scored when 1 a statistically significant sex difference of the same sign was found in the same anatomical region in a TIV-adjusted dataset and in the TIV-matched subsample; or 2 when a statistically significant sex difference in a particular brain region was neither found in the TIV-adjusted dataset and in the TIV-matched subsample.

Trying to identify the brain areas where sex differences might have the highest and lowest likelihood of occurring, a replication score was calculated. This calculation was carried out using the results obtained in the TIV-matched subsample, as well as with results from adjusted datasets that proved to be trustworthy.

In a second step, the individual scores for each VOI in the different datasets were added together, and the final score obtained was interpreted without attending to its sign. A difference was considered highly replicable when it was observed in all or all except one of the included data sets. As Figs. These results are highly similar to those from previous studies assessing the total gray matter and local volumes in pre-selected neuroanatomical areas [ 1112131718 ].

Panels left and right present odd and even numbered brain anatomical regions of the AAL atlas, which with the exception of the lobules of the cerebellar vermis are located in the left and right hemisphere, respectively. Effect sizes of sex differences in each dataset. Previous studies have shown that the raw volumes of several brain anatomical structures are directly, but not uniformly related to TIV [ 111517183154 ]. Thus, as exemplified in Fig. More specifically, the percent of variance accounted for by TIV ranged from 9.

The slopes of these VOI-TIV linear relationships also showed wide variation across different brain areas, ranging from 0. More importantly, these results also show that the tighter the TIV-VOI relationship, the larger and more likely the sex differences, thus revealing that differences between females and males in raw gray matter volume are at least partially dependent on TIV scaling effects.

As shown in Fig. Taken together, these results reveal that, in the absence of any effects of sex, a TIV difference of the same magnitude as the one observed in the main sample results in widespread and medium-to-large local volume differences that unfailingly favor the groups with larger TIVs. Local volumetric differences attained p values below 0. Therefore, the low number, the reduced size, and the bidirectionality of the sex differences observed in the TIV-matched subsample is due to removal of TIV effects and not to sex reduced statistical power.

Accordingly, neither the significance levels nor the effect sizes of the sex differences observed in this subsample were correlated Spearman rho 0. This qualitative conclusion was validated by a correlational analysis.

Therefore, most sex differences observed in the raw gray matter volumes of unselected females and sex seem to result from TIV-scaling effects, making it necessary to remove the effects of TIV before evaluating any possible specific sex differences in gray matter volume. As expected, TIV-adjustment reduced the number and size of sex differences in gray matter volume.

However, as described below, the number, size, and direction of these sex differences were strikingly dependent on the method used to correct for the TIV effects. As depicted in Fig. As Fig. In the 12 cases where males had larger VOIs than females, the effect size of the differences ranged from 0. In a different vein, it is worth noting that, whereas sex was only a relevant predictor of 31 VOIs, TIV was a significant predictor in all of the VOIs considered in this study.

Accordingly, the semi-partial correlations corresponding to TIV M 0.

Frequently bought together. Add all three to Cart Add all three to List. These items are shipped from and sold by different sellers. Show details. Ships from and sold by Amazon. Customers who viewed this item also viewed.

Page 1 of 1 Start over Page 1 of 1. Sex Volume 1: Summer of Hard. Joe Casey. Sex Volume 4: Daisy Chains. Matt Fraction. Brian Azzarello. Start reading Sex Vol. Don't have a Kindle? Customer reviews. Top Reviews Most recent Top Reviews. There was a problem filtering reviews right now. Please try again later. Format: Paperback Verified Purchase. After the "will they or won't they" near-cliffhanger of the first volume, we were left with a sense of "What happens next"? I'd like to say that question gets answered in this volume, but this volume and the next actually focuses more on the Armored Saint's former sidekick, who is trying to disrupt the organized crime syndicates in the city.

By the end of this volume and, sadly, the next you'll still be wondering where this is all going. And despite the title and covers, there's really very little that's titillating about this series anymore. As you've probably gathered, at this point I've read volume 3 also, and I'm still waiting for something to happen.

Honestly, my recommendation at this point is to wait until the whole series is done and pick it up at some half-price sale in one big chunk, because these succeeding chapters have just not been worth the wait.

While it is true that Casey doesn't live up the multi-layered complexities of the first book in its mani character, Casey also actually expands the universe from the relatively unlikable Cooke. One would not actually want a whole another of the limbo that Cooke was in as Casey did not make Cooke a particularly likable character.

There is no hook, but the tropes are still being deconstructed. The sex in this volume is more contextual and less titillating, although a real flippant use of male-on-male rape mars it. The character arcs here do seem enlarging and slowing down the plot can be frustrating, but I actually think it is a good sign for the problems of the first volume.

It was even better than Volume 1! Once you are entranced by the world, you will not want to leave. While it's true that the names in this graphic novel series can be swapped out with Batman character names of similar characters and nobody would notice the difference, the story is the meat and potatoes. Additionally, the characters have enhanced depth than the series obvious primary influence material, as the no holds barred approach definitely gives you the full package and can show you more layers to the characters.

Not to be missed! Amazing Series, the storyline will draw you in. Format: Paperback. Heavy on drama and character development, Sex heaps another serving of reality on our platter and feeds it to us cold.

Sound familiar? This method incorporates information from all the participants, and having a similar number of participants in each group sex becomes critical to ensure the reliability of the results [ 16 ].

Moreover, each regression coefficient is associated with a significance level, thus making second-level analyses i. The power-corrected proportion method PCP was recently proposed by Liu et al. This method explicitly assumes that the relationship between the TIV and a VOI is not proportional, but instead follows a power law.

This procedure was initially discussed by Arndt et al. The significance threshold was initially set at 0. To maximize statistical power, no corrections for multiple comparisons were initially introduced, and following recent recommendations of the American Statistics Association [ 36 , 37 ], we focused our analysis on effect sizes rather than p values.

More specifically, in decreasing order according to their expected statistical power, the Benjamini, Krieger and Yekutieli [ 38 ] Benjamini and Hochberg [ 39 ], Holm [ 40 ] and Bonferroni-Dunn [ 41 ] corrections for multiple comparisons were tested.

The percent of overlap denotes the proportion of scores that overlap in two normal distributions which means differ in some magnitude, whereas the percent of superiority denotes the probability that a randomly sampled member of population a will have a score Y a that is higher than the score Y b attained by a randomly sampled member from population b [ 46 ]. Previous studies have shown that in the absence of any correction, the local volumes of particular brain areas are directly related to the TIV [ 15 , 17 , 18 , 29 ].

The presence of this relationship in our own raw data was assessed by performing linear regression analyses relating the TIV and each of the VOIs considered in this study.

TIV-VOI adj relationships provided a first and powerful criterion to evaluate the goodness of the different adjustment methods tested in this study. That is, because the aim of the adjustment methods is to get rid of TIV effects and provide an unadulterated estimation of sex differences, satisfactorily adjusted data should not show the linear TIV-VOI adj relationship predicted for the raw data, and the likelihood or size of sex differences in local gray matter volumes should not be associated with TIV-VOI adj slope values.

In these analyses, p values were used instead of test statistics because the former provide standardized versions of the latter that can be compared across all the adjustment methods and samples used in the present study for a more detailed discussion, see [ 30 ].

To obtain a more detailed comparison with the TIV-matched subsample, we analyzed the relative frequency of coincidental and non-coincidental findings of this criterial subsample and each TIV-adjusted dataset. A coincidental result hit was scored when 1 a statistically significant sex difference of the same sign was found in the same anatomical region in a TIV-adjusted dataset and in the TIV-matched subsample; or 2 when a statistically significant sex difference in a particular brain region was neither found in the TIV-adjusted dataset and in the TIV-matched subsample.

Trying to identify the brain areas where sex differences might have the highest and lowest likelihood of occurring, a replication score was calculated. This calculation was carried out using the results obtained in the TIV-matched subsample, as well as with results from adjusted datasets that proved to be trustworthy.

In a second step, the individual scores for each VOI in the different datasets were added together, and the final score obtained was interpreted without attending to its sign. A difference was considered highly replicable when it was observed in all or all except one of the included data sets. As Figs. These results are highly similar to those from previous studies assessing the total gray matter and local volumes in pre-selected neuroanatomical areas [ 11 , 12 , 13 , 17 , 18 ].

Panels left and right present odd and even numbered brain anatomical regions of the AAL atlas, which with the exception of the lobules of the cerebellar vermis are located in the left and right hemisphere, respectively. Effect sizes of sex differences in each dataset. Previous studies have shown that the raw volumes of several brain anatomical structures are directly, but not uniformly related to TIV [ 11 , 15 , 17 , 18 , 31 , 54 ]. Thus, as exemplified in Fig. More specifically, the percent of variance accounted for by TIV ranged from 9.

The slopes of these VOI-TIV linear relationships also showed wide variation across different brain areas, ranging from 0. More importantly, these results also show that the tighter the TIV-VOI relationship, the larger and more likely the sex differences, thus revealing that differences between females and males in raw gray matter volume are at least partially dependent on TIV scaling effects.

As shown in Fig. Taken together, these results reveal that, in the absence of any effects of sex, a TIV difference of the same magnitude as the one observed in the main sample results in widespread and medium-to-large local volume differences that unfailingly favor the groups with larger TIVs.

Local volumetric differences attained p values below 0. Therefore, the low number, the reduced size, and the bidirectionality of the sex differences observed in the TIV-matched subsample is due to removal of TIV effects and not to its reduced statistical power.

Accordingly, neither the significance levels nor the effect sizes of the sex differences observed in this subsample were correlated Spearman rho 0. This qualitative conclusion was validated by a correlational analysis. Therefore, most sex differences observed in the raw gray matter volumes of unselected females and males seem to result from TIV-scaling effects, making it necessary to remove the effects of TIV before evaluating any possible specific sex differences in gray matter volume.

As expected, TIV-adjustment reduced the number and size of sex differences in gray matter volume. However, as described below, the number, size, and direction of these sex differences were strikingly dependent on the method used to correct for the TIV effects. As depicted in Fig. As Fig. In the 12 cases where males had larger VOIs than females, the effect size of the differences ranged from 0. In a different vein, it is worth noting that, whereas sex was only a relevant predictor of 31 VOIs, TIV was a significant predictor in all of the VOIs considered in this study.

Accordingly, the semi-partial correlations corresponding to TIV M 0. The calculated b parameter varied widely across the different regions of interest range 0.

In 13 cases The anatomical localization of all these sex differences is shown in detail in Fig. In the 9 cases where males had larger residual-adjusted VOIs than females, the effect sizes ranged from 0. Scatterplots and outcomes of linear regression analyses of the raw or VBM8-, proportion-, PCP- or residual-adjusted volumes of the right amygdala right , or the left thalamus left vs.

Similarly, the proportion adjustment method reduced the strength and, in most cases, inverted the direction, but it did not remove all the TIV-VOI adj linear relationships Fig. A remaining and inverted relationship between TIV and proportion-adjusted local gray matter volumes had been previously reported [ 11 , 15 , 19 , 21 ], but its relevance for the number and size of sex differences had not been explored.

These slopes were significantly different from zero in 63 proportion-adjusted VOIs Fig. However, the sex differences observed in PCP- and residual-adjusted data as well as those estimated from covariate regressions are free of any influence of TIV.

Therefore, it might be concluded that, only in the covariate regression-, PCP and residual-corrected datasets, and unbiased estimates of sex effects might be obtained. Comparison with the TIV-matched sub-sample. This score only took into account the outcomes of datasets adjusted with methods that are free of TIV effects the covariate regression-, the PCP-, and the residuals-adjusted datasets.

The d values for these differences ranged between 0. The confidence intervals of the estimated effect sizes were relatively broad, thus indicating that the precision of these estimates is suboptimal.

Moreover, in some cases, confidence intervals included the zero value, which introduces some uncertainty about the reliability of these differences. Therefore, we sought to investigate how several procedures to correct for multiple comparisons affected the number of statistically significant sex effects in each TIV-adjusted dataset, as well as in the raw volumetric data.

This effect was more pronounced in the PCP- and residuals-adjusted datasets, in which even the most liberal correction procedures resulted in levels of significance above 0. A similar decrease was observed in the TIV-matched subsample, although the sex difference observed in the right amygdala retained statistical significance across all the correction procedures. On the other hand, the decline in the number of statistically significant differences was less sharp in the covariate regression—and even less so in the proportion and the VBM8-adjusted datasets.

Moreover, in the VBM8-adjusted dataset, adopting Benjamini-Krieger-Yekeuteli-corrected p values resulted in a larger number of statistically significant differences than when using uncorrected p values a paradoxical effect that is not uncommon in studies involving between-group comparisons of brain structure measures [ 55 ]. Finally, the number of differences observed in the raw dataset was mostly unchanged, and only when using the Bonferroni-Dunn correction, two out of comparisons failed to reach statistical significance.

Effect of different procedures to correct for multiple comparisons on the number of sex differences in raw and TIV-adjusted datasets. The results of the present study allow us to draw three main conclusions.

First, most male-female differences in regional gray matter volumes are due to sex-independent TIV-scaling effects. Second, not all methods currently used to remove TIV variation are equally effective and valid. Thus, choosing an appropriate adjustment procedure becomes a critical methodological decision that should be reported in detail and carefully considered when summarizing the results of different studies.

As a result, these two adjustment methods should probably be abandoned for similar conclusions, see [ 16 , 17 , 20 , 29 , 54 ]. Nevertheless, the higher flexibility of this method might recommend its use in particular circumstances e.

Therefore, choosing one of these three valid methods should be guided more by the sample characteristics, the measures that are available, and the experimental design than by any a priori recommendation for a more comprehensive discussion, see [ 16 , 29 , 30 ]. Third, when TIV effects are properly controlled, sex differences in gray matter volumes seem to be relatively infrequent and small.

However, a precise and definitive answer to the question of how many and how large the sex differences in gray matter volume are cannot be provided. In any case, the question of how many sex differences there are might be considered spurious because statistical significance whether or not a consensual but arbitrary probability threshold is surpassed does not equate to scientific relevance, and because statistical significance and, thereby, the number of differences found is critically dependent on sample size.

Indeed, as recently mentioned in a statement by the American Statistical Association [ 36 ], p values have no inferential content, and they do not measure the size or the importance of a result. For the present study and other similar studies, this means focusing more on the size than on the number of sex differences. Nevertheless, it is worth mentioning that not only in this study, but also in others with larger sample sizes [ 11 , 17 , 19 ], the number of statistically significant sex differences is much lower than the number of sex similarities, especially when adopting a significance level corrected for multiple comparisons Fig.

This is the case because, although effect size measurements are independent from the sample size, the sample size affects the precision of their estimation [ 63 ]. Therefore, it might be argued that the actual effect sizes of the sex differences in cerebral gray matter volumes could be larger than those reported in our study. Indeed, several studies [ 11 , 17 , 19 , 65 ] using valid TIV-adjustment methods in samples larger than ours, estimated effect sizes that were similar, but smaller, than those provided here.

This might be illustrated by using the amygdala volume as an example. Thus, our estimated average d values for the right and left amygdala 0. These results clearly show that, even in the anatomical regions at which the largest sex differences were found, gray matter volumes present an impressive degree of overlap ranging between Accordingly, the probability that a randomly sampled person from one sex will have a larger gray matter volume than a randomly sampled person from the other sex never exceeded the The meaning of this observation is better appreciated by comparing it to the size of the somatic male-female differences such those observed in as height, at which overlap is reduced to Small effects might be meaningful [ 42 , 67 ].

Moreover, effect size interpretation is always dependent on the research context [ 68 ]. Thus, small sex differences such as those observed in the present study might become relevant in the context of psychiatric or neurological disorders, whereas they might be far less relevant in many other research contexts [ 69 , 70 ].

However, whether or not this is the case remains to be demonstrated in future studies. First, it should be noted that we used a convenience sample rather than sampling epidemiological techniques that covered a relatively narrow age range and was mainly composed of university students. Although these characteristics are typical of most volumetric studies in non-clinical populations, they may reduce generalizability to other populations.

Although this approach has less anatomical precision than voxel-based analyses, it was chosen because 1 it defines the VOIs before conducting any data analysis, hence avoiding circularity and SHARKing and contributing to the accurate estimation of effect sizes [ 71 , 72 ]; 2 It reduces the number of between-group comparisons, thus contributing to obtaining an adequate balance between sensitivity and statistical power.

More specifically, after setting the power at 0. In this way, restricting the number of between-group comparisons to predefined VOIs allowed us to detect even small effects while maintaining statistical power at much higher levels than those ordinarily observed in neuroimaging studies [ 64 , 73 ]. However, it should be noted that, although the AAL is probably the most commonly used atlas in MRI studies, this atlas was constructed based on the neuroanatomical characteristics of a single brain [ 33 ], and it also presents other limitations inherent to the use of any predefined template [ 74 ].

The datasets containing the raw and adjusted data used during the current study are available from the corresponding author on reasonable request. American Psychiatric Association. Cautionary statement for forensic use of DSM In: diagnostic and statistical manual of mental disorders; McCarthy MM.

Incorporating sex as a variable in preclinical neuropsychiatric research. Schizophr Bull. Miller VM. Why are sex and gender important to basic physiology and translational and individualized medicine? Am J Physiol Circ Physiol. Maney DL. Just like a circus: the public consumption of sex differences.

Curr Top Behav Neurosci. Gender on the brain: a case study of science communication in the new media environment. PLoS One. The Behavioural inhibition system, anxiety and hippocampal volume in a non-clinical population. Biol Mood Anxiety Disord. Hippocampus and amygdala volumes in a random community-based sample of year olds and their relationship to cognition.

Psychiatry Res - Neuroimaging. MR imaging volumetry of subcortical structures and cerebellar hemispheres in temporal lobe epilepsy. Am J Neuroradiol. Sullivan GM, Feinn R. Using effect size-or why the P value is not enough. J Grad Med Educ [Internet]. The relationship between corpus callosum size and forebrain volume. Cereb Cortex. Size matters: cerebral volume influences sex differences in neuroanatomy. Brain size and grey matter volume in the healthy human brain. The power-proportion method for intracranial volume correction in volumetric imaging analysis.

Front Neurosci. Intracranial volume normalization methods: considerations when investigating gender differences in regional brain volume. Head size, age and gender adjustment in MRI studies: a necessary nuisance?

Sex differences in the adult human brain: evidence from UK biobank participants. Problems with ratio and proportion measures of imaged cerebral structures. Psychiatry Res Neuroimaging. Correction for head size in brain-imaging measurements. Correction for intracranial volume in analysis of whole brain atrophy in multiple sclerosis: the proportion vs. Normalization of cerebral volumes by use of intracranial volume: implications for longitudinal quantitative mr imaging.

Selective aging of the human cerebral cortex observed in vivo: differential vulnerability of the prefrontal gray matter. Cereb Cortex [Internet]. Aging, sexual dimorphism, and hemispheric asymmetry of the cerebral cortex: replicability of regional differences in volume. Neurobiol Aging. Sex differences in neocortical structure and cognitive performance: a surface-based morphometry study. Pakkenberg H, Voigt J. Brain weight of the danes.

Cells Tissues Organs. Unsolved problems in comparing brain sizes in homo sapiens. Brain Cogn. Adjustment for whole brain and cranial size in volumetric brain studies: a review of common adjustment factors and statistical methods. Harv Rev Psychiatry. Statistical adjustments for brain size in volumetric neuroimaging studies: some practical implications in methods. Voevodskaya O. Front Aging Neurosci. Functional imaging; Neuroimage [Internet]. Wassertheil S, Cohen J.

Statistical power analysis for the behavioral sciences. A voxel-based morphometric study of ageing in normal adult human brains.

Am Stat [Internet]. Statistical tests, P values, confidence intervals, and power: a guide to misinterpretations. Eur J Epidemiol. Adaptive linear step-up procedures that control the false discovery rate. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Vinet L, Zhedanov A. A simple sequentially Rjective multiple test procedure. J Phys A Math Theor. Dunn OJ. Multiple comparisons among means. J Am Stat Assoc.

Interpreting effects. In: the essential guide to effect sizes; Effect size estimates: current use, calculations, and interpretation. J Exp Psychol Gen. Lakens D. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs.