Later in the essay, I model the evolution of crop-based knowledge and its application to other crops explicitly. New ideas generated from growing one crop benefit farm operators in producing other crops as well. The more crops have in common, the more benefit farmers obtain from applying knowledge across crops. If knowledge evolves independently across crops, producers are less likely to master the production of a large number of crops. For example, if learning about almond production is independent of learning about strawberry production, the probability that a farmer is knowledgeable about both is small. So, learning will lead to specialization. Specialization can also be manifested as focusing on a subset of crops that are similar in agronomic characteristics because farmers can apply knowledge across these crops. Following the same reasoning, this model has implications for the number of farms. Assuming there is a minimum acreage required for each crop to establish production, farmers will exit production if their optimal land demand is smaller than the crop-specific threshold. A faster learning process results in a larger variation in productivity because farmers have a larger probability to increase their knowledge. If we consider the number of farms that produce a specific crop, a larger variation in productivity means that there are more farms exited from production due to lack of knowledge. If the demand of a crop is fixed or increases more slowly than the evolution of knowledge, more farms will exit and the number of farms will decrease. The model and implications are presented in section 2. Numerical simulations illustrating the effect of demand- and supply-side factors on the equilibrium path of farm structure are included in section 3 and section 4 concludes.Hens housed in conventional cage systems produce the majority of eggs worldwide, square plant pots however in recent years many countries have shifted to alternative production systems, such as cage-free aviaries .

Conventional cages house small groups of about 6-7 hens in cages with about 67-86 square inches of space per bird . Cage-free aviaries, on the other hand, house hens in large flocks with approximately 144 square inches of space per hen . The shift away from conventional cages is due to increasing consumer preference for cage-free eggs, resulting from consumers’ concerns about the welfare of laying hens housed in conventional cages . Consumers perceive hens from cage-free systems to have enhanced welfare when compared to hens in conventional cages, despite many consumers being unaware of the meaning behind egg labels or the differences between different production systems . This public perception has also coincided with several states passing legislation that aims to phase out the use of conventional cages. California’s Proposition 12 mandates that all eggs produced and sold in California are cage-free by 2022. Similar legislation has followed, including bills passed in Colorado , Michigan , Oregon , and Washington . Cage-free systems have numerous welfare benefits when compared to conventional cages, including the opportunity to perform natural behaviors such as wing flapping, dust bathing, perching, and flying . However, cage-free systems also come with drawbacks. Adult hens housed in commercial aviaries are prone to injuries including keel bonefractures, which may occur during collisions with tiers, perches, and other features of the aviary . There is a great need to determine why collisions are so common in cage-free aviary systems and what management solutions can be implemented to reduce injuries in cage-free flocks. Young hens, or pullets, are housed in a rearing system for the first 15 to 18 weeks of life before being moved to their adult laying system. The complexity of the pullet rearing environment and early access to vertical space has been shown to play a role in reducing keel bone fracture prevalence in adult hens .

Gunnarsson et al. proposed that rearing chicks on the floor, without access to perches or platforms, impairs the development of spatial cognition. One possibility is that this impairment to spatial cognition could contribute to failed landings and navigation of commercial aviaries, increasing the incidents of fractures. Many studies since Gunnarsson et al. have continued to provide evidence that the complexity of rearing environment influences performance on spatial cognition tasks . However, none of these studies have specifically looked at the development of depth perception and its potential relation with early exposure to vertical space. A deficit in precision of depth perception could explain the occurrence of failed landings and falls, due to an inability to properly gauge the distance to fly or jump.The visual cliff has been used for the evaluation of depth perception and differential visual depth threshold since its invention by Walk et al. in 1957. It utilizes two depths, a shallow side and a deep side, and can be adapted for a variety of species. The subject is placed between the shallow and deep side, in the center of the table. The shallow side is typically level or about a few centimeters below the starting point of the subject, however, the bottom of the deep side is far below the subject. The deep side is covered with a sheet of plexiglass so the subject can perceive this depth but, unbeknownst to the subject, is not in danger of falling. Many designs employ the use of checkerboard patterns to provide visual perspective, allowing for an easy determination of depth. The behavior of the subject is recorded to determine their ability to differentiate the depths and avoid “falling down” the perceptual precipice created by the deep side of the visual cliff.

The visual cliff is a widely used test of depth perception that has the advantage of involving a clear, straightforward choice: Does the animal move to the shallow or deep side? The test is not physically challenging for the subject and there is no training required. Therefore, many animals can be tested and there are no confounding variables of learning or physical ability. Despite these advantages, the visual cliff is not free of flaws. The plexiglass over the “deep side” of the cliff can provide tactile information about the presence of a barrier if the subject comes in contact with the surface. There is also a potential for reflection on the plexiglass, giving a visual indicator of an additional surface over the “deep side.” If the subject can detect the plexiglass barrier through either visual or tactile information, the illusion of a cliff ceases and the test no longer compares the subjects’ reaction to differential depths. Using the visual cliff paradigm, garden pots square chickens have been found to have excellent depth perception, preferring the shallow side of a visual cliff significantly more than the deep side from as early as one day of age . Additionally, four day old chicks readily jump down a drop off of less than 10 inches to join their companions but hesitate if the drop is more than 16 inches . Chicks demonstrate a 2 inch threshold for differential visual depth, meaning, chicks perceive a difference in depth only when the discrepancy is 2 or more inches . Unlike humans, chickens do not require binocular vision, or stereopsis, to perceive depth. Chicks with monocular vision are able to perceive depth as well as their binocular counterparts, with both groups choosing the shallow over the deep side of the visual cliff significantly more . Although all birds have binocular vision, there is no evidence that birds other than certain birds of prey use stereopsis to acquire information on relative depth . Instead, birds use monocular cues such as motion parallax and interposition to judge relative depth . Walk and Gibson provided evidence that the ability to perceive depth is innate in multiple species. However, certain factors during development, such as light and monocular deprivation, can cause impairments in depth perception . This raises an interesting question: can other differences in visual experience, such as reduced experience with height and depth, alter depth perception abilities?The relationship between spatial cognition and rearing environment in laying hens has been evaluated in previous studies using a variety of tests including the jump test, hole board task, radial maze, and detour paradigm. In order to better understand these tests, the definition of spatial cognition must be addressed. Spatial cognition is defined as a multifaceted topic entailing the perception, processing, and interpretation of objects, space, and movement . Spatial cognition encompasses many different aspects of visual perception such as spatial memory and navigation, determining the orientation of objects, and perceiving depth . Despite this diverse array of topics under this broad term, many researchers outside the field of cognition discuss spatial cognition without specifying which aspect they are investigating. By failing to properly define terms or control for a precise aspect of spatial cognition, researchers can inadvertently measure unintended variables and make inaccurate conclusions. Over the last two decades there has been a great deal of interest in how the complexity of rearing environments may affect the development of spatial cognition skills/abilities of laying hens. Multiple tests have been used to evaluate different aspects of spatial cognition, including: jump test, hole board test, radial maze, and detour task.

Each test has targeted objectives aimed at specific cognitive processes, and comes with advantages and disadvantages.Gunnarsson et al. developed a jump test to evaluate spatial cognition in laying hens, however the specific aspect of spatial cognition being explored was not addressed. The jump test involves placing a feeder on an elevated platform with or without a second, lower platform to assist in access. Pullets’ latency to reach the food reward is measured to assess their ability to navigate elevated structures. Chicks were raised with either early access or late access to perches of varying heights. At 9 weeks old, all birds were placed on perches 2 to 4 times daily to encourage perch use. At 15 weeks all food in the home pen was placed on a 60 cm high tier, which the pullets could access from the ground or via nearby perches. At 16 weeks old, food deprived birds were presented with a feeder on an elevated tier in a testing pen. The height of this tier increased by 40 cm each trial up to 160 cm high and an intermediate tier was provided to aid in access for two of these trials. The time it took for the birds to reach the food after entering the testing pen was recorded and used to measure their success and ease at reaching the elevated food reward. There was no difference in the time that it took birds from different rearing treatments to successfully reach the food on the 40 cm tier. However, there was a significant difference between the treatments when the difficulty of the task increased, with more successes and shorter latencies to access the food from the early access to perch group. The number of birds that successfully reached the tier decreased with increasing difficulty for both rearing treatments. Norman et al. also conducted a jump test with chicks reared from hatching in either a control or enriched treatment. The control treatment had no elevated structures while the enriched treatment had eight wooden perches arranged in an A-frame structure and a ramp leading up to a platform with additional perches. Like the Gunnarsson et al. study, this test also utilized staggered tiers that could be used to reach a reward, however the tiers were opaque and companion chicks were used as a reward instead of food. Chicks were tested at 14- 15 and 28-29 days of age by placing them in a compartment with familiar chicks from their home pen in a mesh holding container. For the first test the companion chicks were on a 20 cm high tier and there were no other tiers present. For the second test, the 20 cm tier remained, however the companion chicks were placed on an additional 40 cm high tier. Norman et al. found that there was a significant difference between age groups on success of reaching the reward, however there was no effect of rearing treatment. In addition, there were no significant differences in the latency to complete the jump test between treatment or age.