""" Generate the emotion-history correlation network figure. Project structure after decompression: project_root/ ├── code/ ├── data/ ├── figures/ # created automatically when scripts are run ├── README.md └── requirements.txt This script reads: data/correlation_network.xlsx The input file should be the combined correlation table generated by: build_correlation_network.py This script applies threshold filtering for network visualization: - Basic emotion to complex emotion / historical-understanding edges: r > 0.30 and p < 0.05 - Complex emotion to historical-understanding edges: r > 0.50 and p < 0.05 This script writes: figures/emotion_history_correlation_network.png """ from pathlib import Path import pandas as pd import networkx as nx import matplotlib.pyplot as plt from matplotlib.lines import Line2D # ========================= # 1. Path settings # ========================= BASE_DIR = Path(__file__).resolve().parents[1] DATA_DIR = BASE_DIR / "data" FIGURES_DIR = BASE_DIR / "figures" FIGURES_DIR.mkdir(exist_ok=True) INPUT_FILE = DATA_DIR / "correlation_network.xlsx" OUTPUT_FILE = FIGURES_DIR / "emotion_history_correlation_network.png" # ========================= # 2. Node categories # ========================= basic_emotions = {"joy", "anger", "surprise", "sadness", "hate"} complex_emotions = { "nostalgia", "frustration_and_accomplishment", "educated", "negative_emotion", "awe_and_reflection", } historical_dimensions = { "historical_authenticity", "perspective_through_player_choices", "multiplicity_and_complexity_of_perspective", "awareness_of_cultural_heritage_protection", "awareness_of_difference_between_games_and_archaeology", } # ========================= # 3. Mapping from subcategories to network nodes # ========================= complex_mapping = { "nostalgia": [ "nostalgia", "cultural_heritage_nostalgia", "personal_memory_nostalgia", "collective_identity_nostalgia", "golden_age_nostalgia", "romanticized_nostalgia", "lost_culture_nostalgia", ], "frustration_and_accomplishment": [ "frustration_and_accomplishment", "failure_related_emotions", "unfair_difficulty_frustration", "exploration_restriction_frustration", "level_challenge_achievement", "exploration_achievement", "character_progression_achievement", "historical_constraint_frustration", ], "educated": [ "educated", "satisfaction_from_learning", "aha_moment_or_epiphany", "immersed_in_historical_learning", "emotionally_moved_by_learning", ], "negative_emotion": [ "negative_emotion", "fear_of_violence_or_war", "rejection_of_imperialism_or_colonialism", "guilt_over_historical_injustice", "frustration_with_narrative_injustice", "sorrow_for_historical_tragedies", ], "awe_and_reflection": [ "awe_and_reflection", "awe_at_the_grandeur_of_civilization", "reflection_on_the_passage_of_time", "reevaluating_personal_historical_perspectives", "respect_for_cultural_diversity", "moral_reflection_on_history", ], } complex_lookup = {} for category, subcols in complex_mapping.items(): for subcol in subcols: complex_lookup[subcol] = category def map_node(raw_node: str): """ Map raw variable names or subcategory names to the aggregated network node names. """ node = str(raw_node).strip() # Basic emotions if node in basic_emotions: return node # Complex emotions and subcategories if node in complex_lookup: return complex_lookup[node] # Historical-understanding dimensions and subcategories if node == "historical_authenticity" or node.startswith("Historical_authenticity_"): return "historical_authenticity" if node == "perspective_through_player_choices" or node.startswith( "Historical_Perspective_through_Player_Choices_" ): return "perspective_through_player_choices" if node == "multiplicity_and_complexity_of_perspective" or node.startswith( "Multiplicity_and_Complexity_of_Historical_Perspective_" ): return "multiplicity_and_complexity_of_perspective" if node == "awareness_of_cultural_heritage_protection" or ( node.startswith("Awareness_of_protection_of_culture_heritage_") and node != "Awareness_of_protection_of_culture_heritage_Archaeological_Perspective" ): return "awareness_of_cultural_heritage_protection" if node == "awareness_of_difference_between_games_and_archaeology" or node.startswith( "Awareness_the_difference_of_games_and_archaeology_" ) or node == "Awareness_of_protection_of_culture_heritage_Archaeological_Perspective": return "awareness_of_difference_between_games_and_archaeology" return None def split_comparison(comparison: str): """ Split a comparison string such as 'joy ~ nostalgia' into two raw node names. """ parts = str(comparison).split("~") if len(parts) != 2: return None, None node1 = parts[0].strip() node2 = parts[1].strip() return node1, node2 def classify_edge(node1: str, node2: str): """ Classify edge type according to the aggregated node categories. """ pair = {node1, node2} if pair & basic_emotions and pair & complex_emotions: return "basic_complex" if pair & basic_emotions and pair & historical_dimensions: return "basic_history" if pair & complex_emotions and pair & historical_dimensions: return "complex_history" return "other" # ========================= # 4. Read and filter correlation table # ========================= df = pd.read_excel(INPUT_FILE) df.columns = [str(col).strip() for col in df.columns] required_columns = ["Comparison", "Correlation Coefficient", "p-value"] missing_columns = [col for col in required_columns if col not in df.columns] if missing_columns: raise ValueError(f"Missing required columns in {INPUT_FILE.name}: {missing_columns}") filtered_edges = [] for _, row in df.iterrows(): raw_node1, raw_node2 = split_comparison(row["Comparison"]) if raw_node1 is None or raw_node2 is None: continue node1 = map_node(raw_node1) node2 = map_node(raw_node2) if node1 is None or node2 is None: continue if node1 == node2: continue r = pd.to_numeric(row["Correlation Coefficient"], errors="coerce") p = pd.to_numeric(row["p-value"], errors="coerce") if pd.isna(r) or pd.isna(p): continue edge_type = classify_edge(node1, node2) if edge_type in {"basic_complex", "basic_history"}: threshold = 0.30 elif edge_type == "complex_history": threshold = 0.50 else: continue # Positive correlations only, consistent with the interpretation in the manuscript. if r > threshold and p < 0.05: filtered_edges.append( { "node1": node1, "node2": node2, "Correlation Coefficient": r, "p-value": p, "edge_type": edge_type, "game": row["game"] if "game" in df.columns else "combined", } ) edge_df = pd.DataFrame( filtered_edges, columns=[ "node1", "node2", "Correlation Coefficient", "p-value", "edge_type", "game", ], ) if edge_df.empty: raise ValueError( "No edges passed the filtering thresholds. " "Please check the correlation table, node mappings, or threshold values." ) # If multiple subcategory-level correlations map to the same aggregated edge, # keep the strongest positive correlation. edge_df = edge_df.sort_values("Correlation Coefficient", ascending=False) edge_df = edge_df.drop_duplicates(subset=["node1", "node2"], keep="first") print(f"Number of retained network edges: {len(edge_df)}") # ========================= # 5. Build network # ========================= G = nx.Graph() for _, row in edge_df.iterrows(): G.add_edge( row["node1"], row["node2"], weight=row["Correlation Coefficient"], p_value=row["p-value"], edge_type=row["edge_type"], game=row["game"], ) # ========================= # 6. Label wrapping # ========================= label_map = { "awareness_of_difference_between_games_and_archaeology": "awareness_of_difference\nbetween_games_and_archaeology", "awareness_of_cultural_heritage_protection": "awareness_of_cultural\nheritage_protection", "multiplicity_and_complexity_of_perspective": "multiplicity_and_complexity\nof_perspective", "perspective_through_player_choices": "perspective_through\nplayer_choices", "historical_authenticity": "historical_authenticity", "frustration_and_accomplishment": "frustration_and\naccomplishment", "negative_emotion": "negative_emotion", "awe_and_reflection": "awe_and_reflection", } labels = {node: label_map.get(node, node) for node in G.nodes()} # ========================= # 7. Colors # ========================= color_base = "#FDB462" # Basic emotions color_complex = "#66C2A5" # Complex emotions color_hist = "#8E6CAB" # Historical understanding color_default = "#BDBDBD" node_colors = [] for node in G.nodes(): if node in basic_emotions: node_colors.append(color_base) elif node in complex_emotions: node_colors.append(color_complex) elif node in historical_dimensions: node_colors.append(color_hist) else: node_colors.append(color_default) # ========================= # 8. Node size by weighted degree # ========================= weighted_degrees = dict(G.degree(weight="weight")) node_sizes = [1200 + weighted_degrees[node] * 900 for node in G.nodes()] # ========================= # 9. Edge width by correlation coefficient # ========================= edges = list(G.edges(data=True)) edge_widths = [1 + d["weight"] * 6 for _, _, d in edges] # ========================= # 10. Layout # ========================= pos = nx.kamada_kawai_layout(G, weight="weight") manual_shift = { "hate": (-0.18, 0.00), "joy": (0.14, -0.10), "frustration_and_accomplishment": (0.10, -0.06), "historical_authenticity": (0.12, -0.01), "awareness_of_difference_between_games_and_archaeology": (-0.05, 0.02), "perspective_through_player_choices": (-0.02, 0.06), "multiplicity_and_complexity_of_perspective": (0.08, 0.03), } for node, (dx, dy) in manual_shift.items(): if node in pos: pos[node][0] += dx pos[node][1] += dy # ========================= # 11. Label positions # ========================= label_pos = {} for node, (x, y) in pos.items(): if node in historical_dimensions: label_pos[node] = (x, y + 0.03) elif node in complex_emotions: label_pos[node] = (x, y - 0.03) else: label_pos[node] = (x, y + 0.025) # ========================= # 12. Plot # ========================= plt.figure(figsize=(16, 10), facecolor="white") ax = plt.gca() ax.set_facecolor("white") nx.draw_networkx_edges( G, pos, width=edge_widths, edge_color="#8F8F8F", alpha=0.60, ) nx.draw_networkx_nodes( G, pos, node_color=node_colors, node_size=node_sizes, alpha=0.95, edgecolors="white", linewidths=1.8, ) nx.draw_networkx_labels( G, label_pos, labels=labels, font_size=11, font_color="black", font_family="sans-serif", bbox=dict( facecolor="white", edgecolor="none", alpha=0.75, boxstyle="round,pad=0.15", ), ) # ========================= # 13. Legend # ========================= legend_elements = [ Line2D( [0], [0], marker="o", color="w", label="Basic emotions", markerfacecolor=color_base, markersize=12, ), Line2D( [0], [0], marker="o", color="w", label="Complex emotions", markerfacecolor=color_complex, markersize=12, ), Line2D( [0], [0], marker="o", color="w", label="Historical understanding", markerfacecolor=color_hist, markersize=12, ), ] plt.legend( handles=legend_elements, loc="upper left", frameon=False, fontsize=11, ) # ========================= # 14. Title and save # ========================= plt.title( "Emotion-History Understanding Correlation Network", fontsize=18, pad=18, ) plt.axis("off") plt.margins(0.20) plt.tight_layout() plt.savefig(OUTPUT_FILE, dpi=600, bbox_inches="tight") plt.close() print(f"Figure saved to: {OUTPUT_FILE}")