{ "metadata": { "name": "hw7-Animal Foraging and the Evolution of Goal-Directed Cognition" }, "nbformat": 3, "nbformat_minor": 0, "worksheets": [ { "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "#Animal Foraging and the Evolution of Goal-Directed Cognition\n", "\n", "###Materials by Randal S. Olson\n", "\n", "All code in this notebook is usable under the terms of the BSD license.\n", "\n", "This entire notebook is licensed under a Creative Commons Attribution 3.0 Unported License.\n", "\n", "\"Creative\n", "\n", " 
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In this 2006 paper, Thomas Hills explored the evolution of an optimal foraging behavior called area-restricted search (ARS):\n", "\n", ">ARS is characterized by a concentration of searching effort around areas where rewards in\n", "the form of specific resources have been found in the past. When resources are encountered\n", "less frequently, behavior changes such that the search becomes less sensitive, but covers more\n", "space. In an ecological context, animals using ARS will turn more frequently following an\n", "encounter with food, restricting their search around the area where food was last encountered. As\n", "the period of time since the last food encounter increases, animals turn less frequently and\n", "begin to move away, following a more linear path. Fig. 1 allows you to experience ARS in a visual\n", "search task (read the directions please).\n", "\n", "
\n", "\n", "You can read more about ARS and Hills' work in the paper. (Free to download: \n", "[http://www2.warwick.ac.uk/fac/sci/psych/people/thills/thills/hills-forcog.pdf](http://www2.warwick.ac.uk/fac/sci/psych/people/thills/thills/hills-forcog.pdf)). \n", "Here, we're interested in the model that Hills used to explore the evolutionary\n", "feasibility of ARS in various environments:\n", " \n", ">To further test the evolutionary robustness of this behavior, I developed a genetic algorithm, using NetLogo, that allows agents to search for food in a two-dimensional space. The goal of the genetic algorithm is twofold: to understand the conditions where ARS is likely to evolve and also to understand a minimal set of mechanistic parameters required for the behavior\u2019s evolution. In this simulation, the agents have three genes that control turning angle per time step when they are \u201con food,\u201d turning angle per time step when they are \u201coff food,\u201d and a memory depth that describes the number of time steps it takes for the animals to progress from the on-food to the off-food turning angle once they have left food (the inverse of memory depth is the slope of change in turning angle per time step). The initial population is generated with a random 24-bit genome per individual (8 bits per gene). The genes assign the foraging rules each generation. Animals then compete in a two-dimensional foraging arena. After an appropriate life span, individuals mate and recombine\u2014disproportionately according to how well they foraged\u2014and a new generation is created, subject to a small mutation rate.\n", " \n", "Following good practice, Hills also included a link to an implementation of the model: [http://ccl.northwestern.edu/netlogo/models/community/ARS-Genetics](http://ccl.northwestern.edu/netlogo/models/community/ARS-Genetics)\n", "\n", "Now, let's say we're animal behavior researchers who specialize in ARS and we want to explore Hills' findings a little more in his model. Unfortunately, NetLogo doesn't work on our computer, so we have the re-implement the model ourselves. Below is one such implementation of Hills' model in Python using IPythonBlocks.\n", "\n", "##Homework\n", " \n", "1) Read through the Python code and write comments taking note of where Hills' description matches the code. Does this implementation do anything differently than described in Hills' text?\n", "\n", "2) Run the model for 100 generations and observe how the forager behavior evolves over that time period. Does ARS evolve? How quickly does it evolve?\n", "\n", "3) Run the model a few more times for 100 generations each and observe how the forager behavior evolves over that time period. Does ARS evolve? Does it evolve faster or slower than the other runs?\n", "\n", "4) Change the number of food patches, `num_patches`, and size of the food patches, `patch_size`, and run the model again. Does ARS evolve with those settings? Repeat this procedure several times. Can you find settings that cause ARS to no longer evolve?\n", "\n", "5) It's important to run evolutionary models several times to make sure that the phenomenon you observe isn't simply due to chance. Fortunately, it's easy to repeat evolutionary models several times in replicate in a short amount of time. Modify the Python code to run the model with default settings in replicate 30 times, save the data from each run separately, and plot the mean of the fitness and traits for those 30 replicates over all 100 generations with 95% confidence intervals.\n", "\n", "6) Repeat #5 except with settings that do not select for the evolution of ARS behavior. Plot the mean and 95% confidence interval of the fitness and traits of these runs on the same plot as #5. What traits are different between the two experiments? Why?\n", "\n", "7) What do these findings entail in biological terms?\n", "\n", "8) What are the limitations of this model?\n", "\n", "9) What are some possible extensions of this model? What other questions could be explored with this model? We will use these ideas for the next part of the homework.\n", "\n", "Bonus) Find a quantitative measure that captures how much the forager follows an area-restricted search. Plot it over evolutionary time for an experiment that evolves ARS." ] }, { "cell_type": "code", "collapsed": false, "input": [ "# load all of the libraries we'll need for the model\n", "from ipythonblocks import BlockGrid\n", "from IPython.display import clear_output\n", "import random" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 12 }, { "cell_type": "code", "collapsed": false, "input": [ "# environment variables\n", "world_width = 150\n", "world_height = 150\n", "num_patches = 20\n", "patch_size = 5\n", "\n", "# genetic algorithm variables\n", "population_size = 30\n", "mutation_rate = 0.01\n", "generations = 100\n", "evaluation_ticks = 1000\n", "number_evaluations_per_generation = 1\n", "visualization_display_frequency = 1" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 13 }, { "cell_type": "code", "collapsed": false, "input": [ "# load all of the functions we'll need for the model\n", "\n", "def setup_world():\n", " \"\"\"\n", " Sets up a grid world with randomly-placed food patches\n", " and returns the grid.\n", " \"\"\"\n", " \n", " # setup world visualization\n", " grid = BlockGrid(world_width, world_height, block_size=4)\n", " grid.lines_on = False\n", " \n", " # choose random (x, y) positions to place food patches\n", " patch_centers = []\n", " \n", " for i in range(num_patches):\n", " patch_x = random.randrange(0, world_width)\n", " patch_y = random.randrange(0, world_height)\n", " \n", " patch_centers.append((patch_x, patch_y))\n", " \n", " # mark the centers of the food patches\n", " for (x, y) in patch_centers:\n", " grid[x, y].green = 255\n", " \n", " # fill in the food patches\n", " for iteration in range(patch_size):\n", " blocks_to_fill = []\n", " \n", " for x in range(world_width):\n", " for y in range(world_height):\n", " # if any of the non-diagonal blocks next to this block have food\n", " if (grid[(x - 1) % world_width, y].green == 255 or\n", " grid[x, (y - 1) % world_height].green == 255 or\n", " grid[(x + 1) % world_width, y].green == 255 or\n", " grid[x, (y + 1) % world_height].green == 255):\n", " blocks_to_fill.append([x, y])\n", " \n", " for (x, y) in blocks_to_fill:\n", " grid[x, y].green = 255\n", " \n", " return grid\n", "\n", "def convert_bit_list_to_int(bit_list):\n", " \"\"\"\n", " Takes a list of bits (binary representation of a number) and converts it into the corresponding integer.\n", " \"\"\"\n", " return int(str(bit_list).strip(\"[]\").replace(\", \", \"\"), 2)\n", "\n", "def fitness_evaluation(world, genotype, display=False):\n", " \"\"\"\n", " Evaluates the given genome of a forager and returns its fitness.\n", " \"\"\"\n", " \n", " # map the genotype into its corresponding phenotype\n", " # first 8 bits encode the angle the forager turns when it doesn't find any food on the current grid\n", " forager_turn_no_food = convert_bit_list_to_int(genotype[0:8])\n", " \n", " # next 8 bits encode the angle the forager turns when it finds food on the current grid\n", " forager_turn_food = convert_bit_list_to_int(genotype[8:16])\n", " \n", " # last 8 bits encode how many ticks the forager stays in the \"food\" state when it doesn't find any food\n", " # on the current grid before moving into the \"no food\" state\n", " forager_memory = convert_bit_list_to_int(genotype[16:24])\n", " \n", " # forager starting location and facing\n", " forager_x = random.randrange(0, world_width)\n", " forager_y = random.randrange(0, world_height)\n", " forager_facing = random.randrange(0, 360)\n", " \n", " ticks_since_last_food = forager_memory\n", " \n", " for tick in range(evaluation_ticks):\n", " ticks_since_last_food += 1\n", " \n", " # if food is present at current location OR memory limit has not been exceeded yet\n", " # block.green == 255 and block.red == 0 means there is food and the predator hasn't eaten it\n", " # block.green == 255 and block.red == 255 means there is food and the predator has eaten it\n", " # block.green == 0 and block.red == 0 means there is no food and the predator hasn't visited that block\n", " # block.green == 0 and block.red == 255 means there is no food and the predator has visited that block\n", " if (world[forager_x, forager_y].green == 255 and world[forager_x, forager_y].red == 0):\n", " ticks_since_last_food = 0\n", " \n", " # mark the location where the predator has visited\n", " world[forager_x, forager_y].red = 255\n", " \n", " # forager is in \"food\" mode\n", " if ticks_since_last_food < forager_memory:\n", " forager_facing = random.triangular(forager_facing - 2.0 * forager_turn_food, forager_facing + 2.0 * forager_turn_food)\n", " # forager is in \"no food\" mode\n", " else:\n", " forager_facing = random.triangular(forager_facing - 2.0 * forager_turn_no_food, forager_facing + 2.0 * forager_turn_no_food)\n", "\n", " forager_facing %= 360\n", " \n", " # move the forager in the direction it's facing (diagonal moves allowed)\n", " direction_to_move = int(((forager_facing + 22.5) % 360.0) / 45.0)\n", " \n", " offset_dict = { 0:(1, 0), # to the right\n", " 1:(1, 1), # up and to the right\n", " 2:(0, 1), # up\n", " 3:(-1, 1), # up and to the left\n", " 4:(-1, 0), # to the left\n", " 5:(-1, -1), # down and to the left\n", " 6:(0, -1), # down\n", " 7:(1, -1) } # down and to the right\n", " \n", " xoff, yoff = offset_dict[direction_to_move]\n", " \n", " forager_x += xoff\n", " forager_y += yoff\n", " \n", " # keep the forager within the bounds of the world\n", " forager_x %= world_width\n", " forager_y %= world_height\n", " \n", " # display the forager's activity if the display flag is active\n", " if display:\n", " clear_output()\n", " world.show()\n", " print(\"forager_turn_no_food: \" + str(forager_turn_no_food))\n", " print(\"forager_turn_food: \" + str(forager_turn_food))\n", " print(\"forager_memory: \" + str(forager_memory))\n", " \n", " # evaluate the fitness by counting how many food blocks the forager picked up\n", " fitness = 0\n", " for block in world:\n", " if block.red == 255 and block.green == 255:\n", " fitness += 1\n", " \n", " # also unmark the predator's movement to \"clean up\" for the next evaluation\n", " block.red = 0\n", " \n", " # disallow <= 0 fitness\n", " if fitness <= 0:\n", " fitness = 1\n", " \n", " return fitness" ], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 14 }, { "cell_type": "code", "collapsed": false, "input": [ "# run the genetic algorithm to evolve the prey foragers\n", "\n", "# best forager data\n", "best_fitness_over_time = []\n", "best_forager_turn_no_food_over_time = []\n", "best_forager_turn_food_over_time = []\n", "best_forager_memory_over_time = []\n", "\n", "# population averages\n", "population_fitness_over_time = []\n", "population_turn_no_food_over_time = []\n", "population_turn_food_over_time = []\n", "population_memory_over_time = []\n", "\n", "# seed the GA populaton with blank genotypes\n", "population = []\n", "for forager in range(population_size):\n", " genotype = []\n", " for bit in range(24):\n", " genotype.append(0)\n", " \n", " population.append(genotype)\n", " \n", "for generation in range(1, generations + 1):\n", " # evaluate the fitnesses of the population of foragers\n", " world = setup_world()\n", " population_fitnesses = []\n", " highest_fitness = 0\n", " fittest_genotype = None\n", " for genotype in population:\n", " fitness = 0\n", " for evaluation_number in range(number_evaluations_per_generation):\n", " fitness += fitness_evaluation(world, genotype) / float(number_evaluations_per_generation)\n", " \n", " population_fitnesses.append(fitness)\n", " \n", " if fitness > highest_fitness:\n", " highest_fitness = fitness\n", " fittest_genotype = list(genotype)\n", " \n", " # keep track of best forager over time\n", " best_fitness_over_time.append(highest_fitness)\n", " best_forager_turn_no_food_over_time.append(convert_bit_list_to_int(fittest_genotype[0:8]))\n", " best_forager_turn_food_over_time.append(convert_bit_list_to_int(fittest_genotype[8:16]))\n", " best_forager_memory_over_time.append(convert_bit_list_to_int(fittest_genotype[16:24]))\n", " \n", " # keep track of population averages over time\n", " population_fitness = 0.0\n", " population_turn_no_food = 0.0\n", " population_turn_food = 0.0\n", " population_memory = 0.0\n", " \n", " for index, genotype in enumerate(population):\n", " population_fitness += population_fitnesses[index]\n", " population_turn_no_food += convert_bit_list_to_int(genotype[0:8])\n", " population_turn_food += convert_bit_list_to_int(genotype[8:16])\n", " population_memory += convert_bit_list_to_int(genotype[16:24])\n", " \n", " population_fitness_over_time.append(population_fitness / float(population_size))\n", " population_turn_no_food_over_time.append(population_turn_no_food / float(population_size))\n", " population_turn_food_over_time.append(population_turn_food / float(population_size))\n", " population_memory_over_time.append(population_memory / float(population_size))\n", " \n", " # display the performance of the fittest forager at a regular interval\n", " if generation % visualization_display_frequency == 0:\n", " fitness_evaluation(world, fittest_genotype, display=True)\n", " print(\"generation \" + str(generation) + \" fittest genotype\")\n", " \n", " # apply fitness proportionate selection to determine the genotypes that reproduce into the next generation\n", " fitness_sum = sum(population_fitnesses)\n", " new_population = []\n", " \n", " for new_offspring_counter in range(population_size):\n", " roll = random.random() * fitness_sum\n", " sum_so_far = 0\n", " for index, genotype in enumerate(population):\n", " sum_so_far += population_fitnesses[index]\n", " \n", " # if the genotype is chosen to reproduce\n", " if sum_so_far >= roll:\n", " parent = list(genotype)\n", " offspring = []\n", " \n", " # copy the offspring's genotype from the parent's genotype\n", " for bit in parent:\n", " # mutations sometimes occur\n", " if random.random() <= mutation_rate:\n", " offspring.append(random.randrange(0, 2))\n", " else:\n", " offspring.append(bit)\n", " \n", " new_population.append(offspring)\n", " break\n", " \n", " population = list(new_population)" ], "language": "python", "metadata": {}, "outputs": [ { "html": [ "
" ], "output_type": "display_data", "text": [ "" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "forager_turn_no_food: 80" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\n", "forager_turn_food: 0\n", "forager_memory: 0\n", "generation 10 fittest genotype" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\n" ] } ], "prompt_number": 15 }, { "cell_type": "code", "collapsed": false, "input": [ "# run this code after the model finishes to see graphs of the forager's traits evolving over time\n", "%pylab inline\n", "\n", "figure()\n", "title(\"Fitness over time\")\n", "plot(best_fitness_over_time, label=\"Best\")\n", "plot(population_fitness_over_time, label=\"Average\")\n", "legend()\n", "\n", "figure()\n", "title(\"Turn angle (no food) over time\")\n", "plot(best_forager_turn_no_food_over_time, label=\"Best\")\n", "plot(population_turn_no_food_over_time, label=\"Average\")\n", "legend()\n", "\n", "figure()\n", "title(\"Turn angle (w/ food) over time\")\n", "plot(best_forager_turn_food_over_time, label=\"Best\")\n", "plot(population_turn_food_over_time, label=\"Average\")\n", "legend()\n", "\n", "figure()\n", "title(\"Memory length over time\")\n", "plot(best_forager_memory_over_time, label=\"Best\")\n", "plot(population_memory_over_time, label=\"Average\")\n", "legend()" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "\n", "Welcome to pylab, a matplotlib-based Python environment [backend: module://IPython.kernel.zmq.pylab.backend_inline].\n", "For more information, type 'help(pylab)'.\n" ] }, { "output_type": "pyout", "prompt_number": 16, "text": [ "" ] }, { "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n", "text": [ "" ] }, { "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAEICAYAAACQzXX2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n", "text": [ "" ] }, { "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREtQQieEEC1BCZ0QQrQEJXRCCNES/x8KvCBSkPuhxwAAAABJRU5ErkJggg==\n", "text": [ "" ] }, { "output_type": "display_data", "png": "iVBORw0KGgoAAAANSUhEUgAAAWgAAAEICAYAAAByEW6PAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nS4ncAAAAAElFTkSuQmCC\n", "text": [ "" ] } ], "prompt_number": 16 }, { "cell_type": "code", "collapsed": false, "input": [], "language": "python", "metadata": {}, "outputs": [], "prompt_number": 16 } ], "metadata": {} } ] }