Experimental subjects were 16 Winstar strain rats, 8 males (ME1, ME2, ME3, ME4, ME5 ME6, ME7, and M8) and 8 females (HE1, HE2, HE3, HE4, HE5, HE6, HE7, and HE8). Another 16 rats were control subjects (8 males: MC1, MC2, MC3, MC4, MC5, MC6, MC7, and MC8, and 8 females: HC1, HC2, HC3, HC4, HC5, HC6, HC7, and HC8). At the beginning of the experiment, female rats weighed 279 ± 14.7 g average and were 90 days old. Male rats weighed 413.85 ± 12.65 g (means ± SEM) and had the same age. Animals were singly housed in standard individual cages in an equipped room (21C +/− 2C) under a 12:12 light-dark cycle. All procedures in the present study were performed in accordance to the principles outlined by the Mexican Official Norm (NOM-062-ZOO-1999), Technical Specifications for the Production, Care and Use of Laboratory Animals.

All experimental operations were performed in individual home cages, constructed of Plexiglas (13 × 27 × 38 cm), with sawdust on the bottom, which was changed every 3 days. We used a precision electronic scale for body weight and food consumption measurements.

Standard Purina Roden Chow food for laboratory animals was used. Food consumption was registered in grams/ day and water consumption in ml/day.

Rats were divided in two groups, experimental and control, equated by body weight. Experimental group (males n = 8, females n = 8) initiated with a 15 days baseline period to determine body weight, food and water intake. Food was removed the following morning at 9 h during 72 hours to initiate the first deprivation period. Subjects returned to free access conditions during a 15 days period. This food deprivation-free access cycle was repeated until completing three cycles (periods 1, 2 and 3). The lasted free access period lasted 30 days. Control group (males n = 8, females n = 8) had free access to food and water during all the experiment. Food and water intake, and body weight were measured daily.

Data (food and water intake and body weight) was registered every day at 9 h in the morning, during 84 days. After data collection, we calculated means ± and standard error for body weight, food and water consumption, cumulative food intake and growth rate for both groups. Growth rate (Gr) was obtained as follows:

where fw represents the final weight and iw initial weight in a free access period; and t represents this period duration (in this case it was of 15 days). The last post-deprivation period was subdivided in two; from this subdivision periods 4 and 5 were obtained. A parametric contrast method, the Student’s t test was used to compare each group averages; since two independent groups of individuals were observed regarding a numerical variable under the assumptions of normality and equal variance. The graphics were created with Sigma Plot^®.

During the first 15 days there was no significant difference in the animals baseline weights (Figure 1). During the 72 h of the first food deprivation period, weight registered in experimental males was 375 ± 12 g (means ± SEM) versus control males 431.3 ± 13.2 g (means ± SEM) and experimental females 245 ± 13.2 g (means ± SEM) versus control females 286.1 ± 13.7 g (means ± SEM). Statistical difference was significant p < 0.01. In the post-deprivation period, weight loss was recovered. It took experimental males 15 days to reduce body weight difference to 443 ± 13.1 g (means ± SEM) compared with control males 454.8 ± 11 g (means ± SEM) P < 0.05, while it took experimental females 5 days to recover weight and show no significant difference with control females. Such body weight loss and recovery repeated itself in the following deprivation-post-depriva- tion cycles. Experimental and control subjects did not show significant differences in body weight at the end of the experiment.

Within the first 15 days total food intake was not statistical different (Figure 1). During deprivation periods, food consumption was 0 g for experimental subjects. During the first day of post-deprivation period, total food intake was significantly larger than during subsequent days (binge eating). The first total food intake was 37.3 ± 0.8 g (means ± SEM) for experimental males versus 28.5 ± 1.1 g (means ± SEM) for control males and experimental females consumed 26.25 ± 1 g (means ± SEM) versus control females 18.7 ± 0.5 g (means ± SEM) p < 0.01. Only experimental females maintained this increase on the total food consumption during 4 or 5 days, returning later to control subjects consumption levels. This food consumption increase repeated itself after every deprivation period.

Similarly to total food consumption, total water consumption during the baseline did not have a sadistically significant difference (Figure 1). During the food deprivation period, a water consumption reduction occurred (self-imposed deprivation). Water intake was 37.5 ± 5.2 ml (means ± SEM) for experimental males versus 56.2 ± 1.8 ml (means ± SEM) for control males, and experimental females drank 22.5 ± 3.6 ml (means ± SEM) versus control females 45 ± 1.8 ml (means ± SEM) p < 0.01. Nevertheless, when returned free access, experimental rats showed a significant increase in water consumption (binge drinking) compared to control subjects. Water intake was 81.8 ± 3.5 ml (means ± SEM) for experimental males versus 57.5 ± 1.6 ml (means ± SEM) for control males, and 68.1 ± 4.6 ml (means ± SEM) for experimental females versus 46.2 ± 1.8 ml (means ± SEM) p < 0.01 for control females. This result repeated itself in all post-deprivation periods.

Throughout the experiment, cumulative food consumption did not present significant differences.

During the first free access period, there was not statistically significant difference (Figure 2). The first free access post-deprivation period, presented statistically significant difference. Growth rate was 2.5 ± 0.2 g/day (means ± SEM) for experimental males versus 1.3 ± 0.12 g/day (means ± SEM) for control males; for experimental females 1.5 ± 0.6 g/day (means ± SEM) versus 0.6 ± 0.06 g/day (means ± SEM) for control females, p < 0.01. Statistical difference tended to get reduced; at the end of the experiment growth rate was .22 ± 0.11 g/day (means ± SEM) on experimental males versus 0.69 ± 0.13 g/day (means ± SEM) for control males, p < 0.05. Growth rate in experimental females versus control females was not significantly different during the last free access period.

The relationship between body weight recovery and growth rate was an interesting finding. When body weight decreased during deprivation periods and recovering on the 2nd, 3rd, and 4th post-deprivation periods (free access) growth rate increased. However, during the 5th post-deprivation period, body weight recovery was related to a growth rate decrease (Table 1). This relation was clearer during the last 15-day period, in which statistical difference disappeared in females and significantly diminished in males.

Results show an analysis of the data on post-deprivation periods. Means ± SEM of body weight and growth rate of experimental versus control groups was compared. Arrows indicate the relation between body weight and growth rate of experimental subjects.

Previous weight regulation and compensatory growth studies have used models which expose animals to food deprivation programs. These studies outcomes have not taken post-deprivation growth rate increase into consideration [29] - [33] . In contrast, this experiment examined weight recovery associated with growth rate increase. Data indicated that growth rate increase was inversely proportional to body weight recovery during post-depri- vation periods. Statistical analyses indicated that the 15 days post-deprivation periods were sufficient to detect differences between groups. It is important to indicate that during the last 15 days of the experiment, statistical differences between groups disappeared. These data suggests that deprivation effect on post-deprivation periods has a limited duration, and confirm similar observations that have led some investigators to postulate that growth rate works as a compensatory mechanism, that regulates weight changes to maintain body weight [29] [34] [35] . Table 1 shows post-deprivation period growth rate data. Such information allows to conclude that weight recovery might be regulated by growth rate and does not exclusively depend on food consumption. Also the organism’s physical activity during the post-deprivation periods is a factor that can possibly modify growth rate; however, other experiments are necessary to explore which factors influence growth rate.

The present experiment results support the findings of previous studies showing that deprivation increases food intake during post-deprivation periods [36] -[39] . Several studies have considered that food restriction and binge eating co-occur [36] [37] [40] . Our results demonstrated that this relation appeared during all deprivation-post- deprivation cycles (Figure 1). Statistical analyses indicated that difference in food consumption between experimental and control groups was statistically significant (p < 0.01) during the first 2 days in males and the first 4 - 6 days in females during post-deprivation periods. These data suggest that binge eating might be a regulatory behavior that appears only during the first days of the post-deprivation period. This is supported by total food intake analyses. No statistical differences in accumulative food intake between experimental and control group was observed, suggesting that the general food consumption pattern was not modified. This result supports the alternative explanation of binge eating as a regulatory behavior.

Polydipsia is a stereotyped pattern of drinking that can be caused by a variety of pathological or experimental factors [41] [42] . In order to understand the water consumption increase causes, several types of animal polydipsia models have been developed: a) traumatic (e.g., supraoptic nucleus lesions in the hypothalamus) [43] ; b) pathological (e.g., diabetes) [44] [45] ; c) schedule-induced polydipsia, [46] [47] ; and d) drugs-induced polydipsia [48] [49] . The basic characteristic of polydipsia is its temporary persistence maintained by: programs, drugs or diseases. Our results show a different type of increase in water consumption. The term “binge drinking” is used to refer to post-deprivation water over consumption, due to its similarities with binge eating. Statistical analyses indicated that water consumption between experimental and control groups was statistically significant (p < 0.01) only during the first day of post-deprivation period (binge drinking) in males and females. This data suggests that self-imposed water deprivation, food deprivation and binge drinking co-occur, as with binge eating (Figure 1).

These results suggest that, the deprivation program had the same effect in females and males, that is to say, no differences were observed in the body weight recovery, binge eating or binge drinking. Sefcikova and Mozes [15] reported that adult female rats showed overfeeding episodes when they were exposed to nutrients restriction periods previously. They concluded that a relation between the experience with restrictive diet and food consumption changes include overfeeding. On the other hand, Kennedy [50] indicated that females and males rats show differences in their body weight regulation, because males are more responsive to feeding signals than females. Nevertheless, results obtained on the present experiment did not show differences in body weight recovery between females and males in the post-deprivation periods. The exposition to other types of foods, could possibly determine some type of differences in body weight and food consumption between males and females. For example, it has been observed that female rats consume more sweetened solutions than males, especially when those contain calories [51] [52] . Apparently, sex can represent an important variable on eating behavior study.

Binge eating and binge drinking inadequate operational characterization was a limitation for this study. Scientific evidence has identified these two types of behavior, from an increase of food and water consumption compared to baseline [13] [38] [39] [53] . However, exactly determining the increment amount or percentage that characterizes this type of behavior is still a pending task. Another limitation was the 24 hour consumption measurement, because to identify the time and conditions under which binge eating and binge drinking are presented, is necessary to perform a day long analysis.